Also every post on this forum is a cogent example of top-down causation. Question is, do we find such causation in inanimate nature. — Querius

There are many examples in physics. George Ellis (responding to Sean Carroll) provides an example in the comment section of this post on emergence by Massimo Pigliucci:

"However this billiard ball point of view, based in Newtonian physics, is invalid once one takes quantum physics into account. A classic example is the fact that the mechanism of superconductivity cannot be derived in a purely bottom up way, as emphasized strongly by Bob Laughlin in his Nobel prize lecture, see R B Laughlin (1999): `Fractional Quantisation'. Reviews of Modern Physics 71: 863-874. The reason is that existence of the Cooper pairs necessary for superconductivity is contingent on the nature of the relevant ion lattice; they would not exist without this emergent structure, which is at a higher level of description than that of the pairs. Hence their very existence is the result of a top-down influence from this lattice structure to the level of the Cooper pairs. The concept of a given set of unchanging interacting particles is simply invalid. They only exist because of the local physical context. One can also find many examples where the essential nature of the lower level entities is altered by the local context: neutrons in a nucleus and a hydrogen atom incorporated in a water molecule are examples."

(Notice that the neutron example provided by Ellis was also given by Wayfarer recently.)

I take it that you are not attempting to explain fundamental laws with emergence. As such the topic emergentism is irrelevant to our discussion. — Querius

You initial query in the first post of this thread was about "laws of nature", quite generally, and the source of their universality. What makes you think that some laws are fundamental and some aren't? It's only on the assumption of reductive physicalism, and/or some rather strong thesis of supervenience, that some laws are believed to be fundamental in the sense that they would govern everything that happens in the world.

But what distinguishes the laws of physics (or of "fundamental" physics) from other laws of natures (or from normative principles of biology, cognitive sciences or social sciences) may only be that the former focus on rather general features of material constitution while abstracting to some arbitrary degree from formal principles of organization and contingent boundary conditions.

I would have thought 'the placebo effect' provides a cogent example of top-down causation. — Wayfarer

For sure. It's Jaegwon Kim's argument that Bitbol is rehearsing here. Kim's argument also is sketched in the section Argument against non-reductive physicalism from his Wikipedia page. It this this argument from causal closure that Bitbol responds to.

I am not sure how a discussion about emergentism is relevant to fundamental laws of nature. As I have stated before I have no problem with a secondary (emergent) law like ‘every snowflake is 6-sided’, as a direct consequence of fundamental laws. — Querius

If laws that govern phenomena from a variety of empirical domains (e.g. chemical reactions, natural evolution, the actions of human beings, etc.) don't reduce to the laws of physics, or to so called "fundamental" laws of nature, them there is no reason to think that they are a consequence of them. The laws of physics may explain, partially, how the higher level processes are implemented, but they leave open what the higher level laws themselves are.

The higher level laws (or systemic principles of organization) may depend on contingent facts about boundary conditions, or historically contingent facts about the evolution of those entities. The higher level entities can also be governed by general principles that are quite independent from the laws that govern their low level material constituents since those higher level entities are multiply realizable in different sorts of materials or components (e.g. the same software can run on different hardware architectures.) In that case, it's not the lower level laws that determine the higher level laws. At most, they may enable them though providing a contingent form of implementation. Enablement, though, falls short from causal determination.

Sean Carroll in his book ‘The Big Picture: On the Origins of Life, Meaning, and the Universe Itself’, Dutton, 2016 writes: ... — Querius

Sean Carroll objects to the notion of downward causation because he doesn't understand it. He wrongly believes the possibility of downward causation to contradict the causal closure of the micro-physical domain, as if a macroscopic or systemic cause of a micro-physical event entailed a violation of the laws that govern micro-physical interactions. But downward causation doesn't have this consequence. It isn't something queer, magical, or unphysical. Sean Carroll is thus shooting down a strawman notion, though, to be fair, he isn't alone in wrongly portraying downward causation in this manner; so does philosopher Jaegwon Kim. In his paper Downward Causation without Foundations, Michel Bitbol (while discussing Kim's objections) sets up the problem of physical closure thus:

"The first statement is meant to dismiss the idea of strong emergence, according to which the high-level processes are endowed with autonomous causal powers, and with ability to alter the low-level processes. It does so by assuming that for high-level processes to count as causal powers in the fullest sense, and to be able to alter anything significant in the lower level, they must induce a deviation in the laws of the micro-processes. But if this were the case, two common presuppositions of the scientific picture of the world would be denied: (a) the presupposition of nomological closure of the lower micro-physical level, and (b) the presupposition of causal fundamentalism, according to which “macro causal powers supervene on and are determined by micro causal powers” (Bedau 2002, 10). Strong emergence thus apparently amounts to an indefensible variety of ontological dualism."

Bitbol later adresses the problem of physical closure thus:

"No level of organization can claim any privilege for itself, because every such level is defined (or “constituted”) by a certain scale of intervention and observation. Moreover, no absolute meta-observer, no “view from nowhere,” is available to select one pattern of causes at a certain agent-relative level as the “truly efficient” one. This does not threaten the thesis of causal closure of the domain of physics, but only denies it any ontological significance. Causal closure here means only that it is possible to establish a systematic and self-sufficient network of causal connections relative to a single scale of intervention and experimental access, without having recourse to any other scale of intervention and access. This being granted, causal closure of a low level of organization (say the level of micro-physics) is perfectly compatible with the thesis that there are also efficient causes at an upper level of organization."

Scott Sehon, Free Will and Action Explanation: A Non-Causal, Compatibilist Account, OUP, 2016

George Ellis, How Can Physics Underlie the Mind: Top-Down Causation in the Human Context, Springer-Verlag, 2016

Andrea Kern, Sources of Knowledge: On the Concept of a Rational Capacity for Knowledge, HUP, 2017

How a system behaves is dependent not only on its constituent parts but also on the organization of these parts, which creates a causal web/network in which general behavior arises. — darthbarracuda

Also, in many cases (i.e. many ontological domains, including the objects of quantum physics) the very nature and existence of the material parts causally and/or constitutively depends on properties of the system of which they are proper parts. This can include boundary conditions, topological properties of space-time, etc. George Ellis argues for this in his recent book on the topic of top-down causation. This general point also has been argued by Michel Bitbol in some of his papers on emergence and on the foundation of quantum mechanics.

Ellis's and Bitbol's arguments are quite general, not very controversial, and free of the obscure and speculative quantum woo-woo that sometimes permeates the discussions of natural scientists when they turn to the topic of the mind. This ontological dependence of parts on whole also had been argued by John Haugeland in Pattern and Being and in Truth and Rule Following (both collected in his Having Tought: Essays in the Metaphysics of Mind). Haugeland's ontological points are quite general but Ellis's and Bitbol's discussion show that physics, even so called fundamental physics, is no exception and affords no refuge to the reductionist.

Natural laws are the natural extension of a Cartesian epistemologically-oriented metaphysics, one that rejects teleology in favor of mysterious, immutable forces that exist for whatever reason. One of the alternatives would be a rejection of natural laws as such, in favor of a re-instituted teleology based upon threshold dispositions and power networks. — darthbarracuda

Yes. I think there is much truth to this. According to Cartesian epistemology, the world can never be known directly through perception. There is a disconnect between the things that we (seem to) know empirically from our temporally situated perspective -- substances that have fallible powers -- and the fundamental (so called) entities posited by the exact sciences, that are subject to exceptionless laws.

But the exceptionless laws that govern the theoretical entities posited by the exact sciences (i.e. "basic physics" and the special sciences that allegedly reduce to it) are conceived through abstracting away most or the real and relational properties of the entities that we actually encounter in the natural (and human) world and in the laboratory. The focus of physics is the (mere) material constitution of ordinary objects.

The OP quoted Sean Carroll, a theoretical physicist, who believes that the law of physics express invariant relations quite unrelated to the ordinary time asymmetric notion of cause. The ordinary notion of causality applies to entities that have fallible causal powers -- things that act on one another or that we can make some use of. If we consider a homogeneous set of such entities, seek purely mechanistic explanations of their behaviors and abstract away from their intrinsic teological structure, and also from the practical uses that we can make of them (as real materials or artifacts) then we can achieve some explanations about the manner in which some processes are materially implemented inside of them. But we also lose sight of what those entities are and the real powers that they have. Far from being a fundamental science, physics is a very narrow science. It may appear fundamental to a Cartesian metaphysican who isn't concerned with the fact that the entities that it describes can't be disclosed in experience but rather must be abstracted away from it.

A philosopher could be overrated (i.e. generally being ranked too high relative to her peers) and still be underappreciated. I can't think of any philosopher belonging to the tradition who isn't underappreciated. This includes philosophers who I don't like much. It's a matter of contemporary culture. The same goes with classical music and literature. We could possibly make an exception for Ayn Rand, assuming that she merits the title of a philosopher. She likely is both overrated and over-appreciated.

If the language of thought hypothesis (LOT for short) were right, then the content of a thought would just be the content of the mentalese sentence used by the brain to represent it. I don't see this as a departure from representationalism. It is just one way to cash out the idea that thoughts are the representional contents of representations that are materially implemented in the brain in some way or other. It raises the question who is it who is interpreting the mentalese symbols? If the brain is a syntactic engine, then it does no interpretation. So, the LOT hypothesis threatens to collapse into eliminativism about mental content, it seems to me.

I prefer the antirepresentationalist view that rather equates thoughts (e.g. beliefs, judgments, experiences, etc.) as acts of capacities to believe, judge, perceive, etc. To believe that the cat is on the mat, for instance, is thus an act of representing the cat as being on the mat. It is something that an embodied person does, not her brain. It need not be the representational content of anything. The correct ascription of such mental states to a living individual is grounded on the interpretation of her overall animate behavior (including her linguistic behavior), in such a way that it is disclosed as being rational in light of such a broadly consistent set of mental state attributions. This is roughly Davidson's view of radical interpretation (divorced from his anomalous monism).

Elizabeth Anscombe has rightfully been mentioned as a "significant original" philosophical thinker. If we take care to distinguish significance from influence, then many more philosophers, male or female, have achieved great depths of significance in their work in spite of the fact that they have been underappreciated. Among the male philosophers who's work seem to me to have had deep significance and who might deserve much more widespread influence are Gareth Evans, David Wiggins and Maurice Merleau-Ponty. Among the female philosophers are Jennifer Hornsby, Sabina Lovibond and Sarah Broadie. There ought to be many many more who I either don't know or who I (together with Charles Murray) am underappreciating.

Simply put, he is just saying that it is the deliberation of the possible choices or outcomes before a decision is made that affects the beneficiality of the choice. That is not absurd as you say it is and is completely coherent with the determinism that he expounds on. — intrapersona

Yes, he is saying this... often. And when he is saying this, he is effectively endorsing compatibilist conceptions of control and of free choice. The trouble is that Harris also tries to hold on to some version of Van Inwagen's argument for incompatibilism. He indeed is very often explicitly arguing against compatibilism on the basis of such an argument (and also on the basis of Galen Strawson's "basic argument" against moral responsibility), and his endorsement of those arguments is inconsistent with his tacit endorsements of compatibilism. This is why Dennett says that Harris often seems to be compatibilist in all but name (i.e. whenever Harris isn't arguing for the contrary position!) Harris charges against compatibilists, that they are "redefining" free will, just is his attempt to extricate himself from this muddle. Compatibilists at least are consistent in their definitions of freedom and of the ability to choose; Harris himself, while consistently denying the possibility of free will, isn't endorsing consistent arguments.

Just because what one is poised to do already is based on prior events does not negate the necessity of logical thought or reasoning for making a decision. Predestination works just as well if not better with a reasoning mind.

You yourself are entitled to this compatibilist insight, but Harris isn't. You may want to replace "predestination" with "determinism" in the above, though, since Harris claims to be agreeing with Dennett that determinism doesn't entail predestination and that those two theses, as they relate to freedom of action, ought not to be confused.

As for why you think it denies the central insight of compatibilism, I see no evidence to support your opinions here...

You are misreading me. I didn't say that he was denying the central insight of compatibilism. I suggested, on the contrary, that he was tacitly endorsing this central insight while at the same time explicitly arguing for incompatibilism on the basis of arguments that are inconsistent with this insight. Harris is very explicit in his rejection of compatibilism, while being seemingly unaware that he is depriving himself of the the possibility of making his utilitarian argument about consequences that "matter".

I see no reason to distinguish between preferences and reasons, as if they are two completely separate things. — darthbarracuda

I don't view them to be completely separate either. General preferences (which I've previously identified as preferences-1) are general and abstract -- they reflect rational tendencies -- while the reason why one acts (which singles out one's preference-2) are specific and sensitive to the particular circumstances in which one acts. Both of them are normative since both of them are involved in the operation of practical rationality. Someone who is practically rational must be sensitive both the the reasons why someone must generally act in such or such a way in some general class of situations (e.g. keep one's promises, or avoid engaging in harmful or overly risky behaviors) and also to the specific circumstances that bear on the applicability of such general reasons to the particular situation in which one is called to deliberate and act.

Reasons, in my view, are just static preferences. I have a reason to go for a run today, because I want to be in good shape. I may not actually prefer to go for a run (exercise is hard...), but this preference is over-ridden by the reason (preference) to be fit.

If you have both a reason to do A and a reason to do B, while A and B are incompatible actions, this means that such reasons are general rational preferences. We could label such reasons "reasons-1" There is also another sense of "reason" which is your reason to chose one particular option in a particular occasion. Practical rationality is the ability select among potentially conflicting reasons for acting (reasons-1) which one is suited or relevant to the rationally salient features of the particular situation. Failure to distinguish between the reason ("reason-2") why one acts in the way one does, and the reasons one might have to act in other incompatible ways (reasons that are overridden in the particular case) just reproduces the conflation that you were previously making between two sorts of preferences.

Thus we can have a static grouping of preferences (reasons) if we have a static goal - to be fit, to understand the truth, etc. The division between normative reasons and non-normative preferences thus, in my view, cannot be sustained.

Both sorts of reasons, (1) general, or (2) particular/specific, are normative. Neither ones are "static". That is, both are sensitive to the concerns that are rationally salient in the circumstances in which one acts. Reasons (that is reasons-1) just are the general rational considerations that guide one in the first steps of practical deliberation. When things go well, this process culminates in action (or intention) in accordance with the reason one has to favor the chosen course of action all things considered (to the extend that one is rational, which one may be to a quite limited degree).

What does reason accomplish if not goals, and where do goals come from if not preferences? — darthbarracuda

Practical reason is an ability to arbitrate between potentially conflicting goals. Some specific goal may be judged to take precedence over another goal, in a particular practical situation, when there is a good reason for it to do so. Practical reason is the ability to understand such reasons and to be motivated by them to act accordingly. Reasons themselves, unlike raw desires, baby squirrels or meteorites, do not come from anywhere. It's a category mistake to ask where a reason comes from. One can ask where the human ability to reason practically (i.e. to be sensitive to reasons) comes from, but reasons themselves stand on their own. If someone's reason to do, or to believe, something is bad, what is required in order to show this reasons to be bad, and thereby motivate an agent to abandon it, isn't a story about the causal origin of that reason (or the causal origin of the specific goal that it recommends one to act upon), but rather a good counterargument.

This preference #2, is only determined after a choice is made, posteriorly, it describes the choice which has been made. "She choose X, therefore it was her preference". Since free will is directed toward choices which will be made, this preference #2 is irrelevant to the free will/determinism issue. Preference #2 cannot act as a cause, and to introduce this sense of "preference" is to create ambiguity with the possibility of equivocation. — Metaphysician Undercover

Of course, that was exactly my point. The two concepts, however, are frequently run together, and Darthbarracuda seems to be relying on running them together in order for his anti-free-will argument to run through. At the same time the concept of preference that singles out what a person effectively choses to do when faced with a range of alternatives that she has deliberated over is a perfectly good concept that reflects a quite normal use of the word "preference". ("I prefer to order the salad because I am dieting" is consistent with "I much prefer eating apple pie to eating salad"). It is thus quite useful to distinguish this concept of an 'all things considered preference' precisely to avoid the equivocation with the other concept of an antecedent and merely general preference -- or desire, or habit of choice -- that may conflict with an agent's assessment of what it is that she deems that she ought to do, or with what it is that she effectively chooses to do precisely because she judges that she ought to do it.

But whatever it chooses, it must choose. It doesn't make sense to have a strong preference yet pick the route of least preference satisfaction, otherwise what exactly would a preference even be? There's nothing free about the will here. The will must choose a certain route of action depending on how strong the various preferences influencing it are. — darthbarracuda

Your view that every practical choice that we make is governed by "preference" isn't false but it needs to be qualified. There is a liability to run together two different senses of "preference". In the first sense, a preference-1 is an antecedent desire or inclination. It is manifested in an agent's tendency to chose to engage in some sorts of behaviors, to tend to some specific sorts of needs, or to favor the achievement of some specific ends (e.g. enriching herself financially or fulfilling her promises). Preferences of that sort are general dispositions that get manifested in circumstances appropriate to them, and which may, but need not, be irrational. This sense of "preference" (that I here label preference-1) is roughly equivalent to "desire"; but labeling it such isn't very useful since the conflation that I wish to warn against also applies to two distinct senses of "desire".

In the second sense, an agent's prefered-2 action is what this agent effectively choses to do in the specific circumstances in which she is called to deliberate and act. That the two senses of "preference" are different is displayed in the fact that they often conflict with one another, as I had hinted in a previous post. What an agent choses to do, in response to rational considerations and evaluation of the salient features of her practical situation (e.g. her opportunities, obligations, general concerns, etc.) manifests, by definition, her preference-2, but often goes against some of her preferences-1. (The phenomenon of akrasia, or weakness of the will, highlights a further complication regarding the concept of preference-2 that I am leaving on the side for now.)

On a crude Humean conception of practical rationality that has been popular in analytic philosophy (but has found much disfavor in more recent Anscombe inspired philosophy of action), and that may not be entirely fair to Hume, preference-2 -- what an agent effectively "prefers-2", or choses, to do -- always is the result of some intrinsically stronger preference-1 (brute desire, or habit) winning out over other preferences-1 that it potentially conflicts with.

On that view, reason always is the slave of passions, as Hume would say, since its scope of operation is entirely restricted to instrumental deliberation: finding means to achieving antecedently determined ends. But once the distinction is properly kept in mind in between preferences-1 and preferences-2, the Humean slogan admits of two readings: only the first one of which supports the crude conception of practical rationality as being governed by preferences-1 that we are passively straddled with and have no control over. Under the second reading, the always present "passion" or "motivational state" that leads an agent into action can be, and indeed always is, sensitive to reason. Habits can be molded and they can be overridden. Preferences-2 are rational, although Hume avoids qualifying them thus since he himself is attacking a crude intellectualist conception of rationality.

What Hume can thus be justly taken to be objecting to is a crude conception of reason that portrays is to be entirely independent of habit and motivation. The second reading of Hume's dictum is defended and elaborated upon by David Wiggins in his Ethics: Twelve Lectures on the Philosophy of Morality. The distinction that I have stressed between two sorts of preferences (two sorts of "imperatives") has been stressed in a similar spirit by John McDowell in his paper Are Moral Requirements Hypothetical Imperatives?. Although those texts focus on issues of meta-ethics, they are centrally relevant to the philosophy of action and of practical rationality, generally.

Beethoven, the complete string trios, Grumiaux Trio, Phillips, 1997

Regardless of free will or not, we still consider the consequences of our actions... (at least if we're sane). — anonymous66

If Harris were right about our not having free will, our not being responsible for out decisions, and our actions being entirely determined by past circumstances that lay entirely outside of our control, then, while it could still be judged to be sane in a purely medical/psychological sense for one to consider the consequence of one's actions prior to acting, it would be practically irrational to do so, just as it is irrational for us to deliberate about things that we know not to be in our power to affect generally.

It has been pointed out to Harris that if it is true from one's perspective, at any given instant, that what one is poised to do already had been determined at that time by one's (and the Universe's) past history then it is pointless to deliberate what to do. Harris's reply to this seemingly absurd practical consequence of his view is to claim that while we can't control the causes of our action, our actions nevertheless have consequences and since consequences matter we ought to take them into account while deliberating what to do. But this answer is completely point missing and is a garbled attempt to take in stride the central insight from compatibilism while, at the same time, denying the cogency of compatibilism.

If we can't control what it is that we are going to do, then we can't control what the consequences of our actons will be, either. If Harris were consistent he would have to claim that, while the consequences of our action matter greatly, it is a sad tragedy of our human predicament that we are powerless to control what it is that we are going to do and we must therefore just as well stand back -- wait and see what the Universe already had in store for us -- and cherish however little confort we can get from the though that the consequences of our resigned inaction were inevitable and thus leave us blameless.

But what are these reasons, other than preferences (i.e. needs, desires, concerns, etc)? — darthbarracuda

This question involves a category error. Reasons are categorically distinct from preferences (or desires, concerns, etc.). If you need milk and believe the corner store to be open, then you may believe that you have a good reason to go to the corner store. But you may be mistaken about that if the corner store is in fact closed. Your being right or wrong about that need not be conditioned by your preferences. You may want to say that upon learning that the store is closed the acquisition of this knowledge leads you to reorder your preferences (e.g. you now prefer to stay home and drink orange juice). But this reordering of preference of yours is sensitive to the fact that it is pointless to go to the corner store to buy milk when the corner store is closed.

Some philosophers with a Humean bent believe that only instrumental actions (things done for the sake of something else, or for the sake of some end) are sensitive to (instrumental) rationality while the choices of the ends themselves aren't sensitive to reasons. The latter merely reflect brute "passions", which we are passively affected by or straddled with, according to such philosophers. But that is a prejudice. It is routine for human beings to reflect on the cogency of their pursuing the ends that they pursue, and have their wills reoriented on the basis of such deliberation. Many among our ends are rational ends (i.e. sensitive to reasons) rather than them all being brute a-rational desires. This is not to say that our natural or culturally ingrained proclivities don't effect the ends that we take ourselves to be justified to pursue, but we hardly are slaves to those proclivities.

Which is of course true.

For every action there is a preference. The act of choosing one's preferences is an act itself, which requires a preference that was not chosen. — darthbarracuda

This may only seem true from the standpoint of a metaphysics of Humean events that are individuated independently of their relational and causal relationships. This Humean view also tends to assimilate each and every manifestation of a "mental state" (including manifestations of so called "motivational states") with an isolable "event" in space and time. From the point of view of a more Aristotelian metaphysics -- a naturalism that is more biologistic than physicalistic -- acts of the will are actualizations of rational/practical embodied capacities rather than them being self standing events that are merely caused to happen by antecedent events and conditions.

If one drops the Humean view, actions and choices (including endorsements of values) can thus be seen as manifestations of a rational agent's orientation of her will. This is a manifestation of her sensitivity to the reasons that she has to act in this or that way in such or such general range of circumstances. If one seeks to trace further back the causal antecedents of the actualization of such rational/practical capabilities then what one primarily finds aren't more ancient "acts of will" but rather the conditions that led up to, or enabled, the formation of that person's rational character and capabilities. This account need not lead a free will defender to any troublesome regress since there is no need to deny that one must have had the good fortune to have sufficiently matured intellectually (thanks to one's biology and culture) before one can act freely.

I don't think it is quite true that we have no control over our preferences or desires. Many preferences result from habits that we can modify. We can ween ourselves off from addictions (to watching TV or eating too much sugar, say). Secondly, human beings routinely act against their preferences or desires. When our desires and values clash, our desires can sometimes cloud our better judgment and lead us to betray our values, but it also often happens that our values motivate us to act against our "raw" desires (i.e. against what we would want to do if we didn't know any better).

Hard determinists may argue that values just are the same sort of things as preferences and desires: that they are fully predetermined by inborn temperament and past conditionings. But this is to overlook the fact that what values we endorse is sensitive to our present reasons for endorsing them. It is misleading to say that we can't choose what good reasons we have for acting in the way we do (or endorse the values we do). A reasonable account of free will doesn't portray the will as an ability to act, or to chose, in a arbitrary manner unconditioned even by reason, but rather portrays it as an ability to act in accordance with reason.

Local realism is just the union of locality and realism. — Michael

Agreed, though "realism" had a special technical meaning referring to the determinateness of quantum measurements before they are effected. As for "action at a distance" check my short references, especially the last one.

It does support non-locality, as Dr. Henson says: — Michael

Yes it does, but the mention of an "action at a distance" is misleading. Nonlocality has long been seen as a consequence of QM (ever since Niels Bohr replied to the paper by Einstein, Podolski and Rosen) and is commonly seen as having been demonstrated by the empirical verification of the violation of Bell's inequalities. However, this non-locality of QM must be understood merely as the denial of local realism and it doesn't entail "action at a distance" where such a phrase is meant to imply that there is a causal interaction taking place at a speed exceeding the speed of light.

See also the first answer to the question How to understand locality and non-locality in Quantum Mechanics?

So I'm right to be confused that these are all supposed to be axioms of the theory? — Michael

Yes it is confusing because Bell's theorem isn't a theory but rather a statement of incompatibility between a set of assumptions. Physicists who endorse the empirical and theoretical validity of QM seldom endorse assumptions (3) and (4) in Tom's table. (Those rather are consequences of local hidden variable theories that aim at reconciling QM with the intuitive "realist" assumptions of classical physics). So, to suggest that (1) is an "axiom" of QM just because it is an "axiom" of Bell's theorem is not just misleading, but confused.

I'm confused by this. Isn't Bell's theorem supposed to show that 3. and 4. cannot both be true? — Michael

Not quite. It is rather supposed to show that those assumptions are jointly incompatible with the statistical predictions of QM. If you then accept the conclusion of "Bell's Theorem" (i.e. this statement of incompatibility) and wish to save the empirically verified predictions of QM (as you probably should) then you have to question at least one of those assumptions. (And keep in mind that "free will" in Tom's statement of the first "axiom" only refers to the lack of predetermination of the setup of the measurement apparatus.)

1. Freedom of choice. The freedom to choose which experiment to perform independently of the object to be measured. i.e. The Free Will of the Experimenter — tom

You are paying no attention whatsoever to the ongoing argument or to the philosophical issue about free will and determinism. Merely using bold characters doesn't validate your assumption of a crude equivalency between the affirmation of free will and the denial of superdeterminsm. What is assumed in order to derive the validity of Bell's theorem is that superdeterminism is false, not that human beings have free will. But the falsity of determinism just is generally assumed to be part of the standard understanding of QM; and if determinism is false then, a fortiory, so is superdeterminism. Superdeterminism just is an extravagant metaphysical doctrine devised as a loophole in order to save QM together with local hidden variables. What is assumed -- your so called "free will axiom" -- in order for Bell's argument to go through merely is that the setup of the measuring apparatus (e.g. the determination of which of two conjugate variables are being measured) together with the result of the measurement aren't predetermined.

In summary again, (1) since most physicists don't care at all for local hidden variables, they don't care for superdeterminism either. And (2) since the question of the compatibility of free will and determinism is central to the philosophical debate you can't crudely equivocate between "free will" and "indeterminism" without begging the question.

Yeah, this confused me too. I don't speak French, but there is a cognate in Spanish, querer decir, which also means 'mean,' but you commonly use it to ask what a word or piece of language in the abstract means, like ¿qué quiere decir 'caballo?' – 'what does 'caballo' mean?' — The Great Whatever

Yes. When inquiring about the usual meaning of a word or phrase we can sometimes ask "Qu'est-ce que ça signifie?" but it is much more usual to ask "Qu'est-ce que ça veux dire?" This last question would be the standard translation of "What does it mean?"

But then, we can also ask "Qu'est-ce qu'elle veut dire (par là)?" meaning exactly the same as "What does she mean (by that)?" in cases where the communicative intention appears to go beyond, or be more precise than, the mere semantic/conventional meaning.

Yeah, I think you're right and that's why Derrida renders bedeutung as 'vouloir-dire.' — csalisbury

It is a bit unfortunate, though, that 'vouloir dire' is also commonly used, just as the English 'meaning' is used, to refer to the conventional linguistic/semantic meaning of a sign and not merely to refer to the communicative intention of whoever uses this sign, on a particular occasion, in speech or thought/soliloquy.

What are the axioms of Bell's Theorem? — tom

Bell's theorem just is the statement that the statistical predictions of QM are inconsistent with all local hidden variable theories. Since superdeterminism provides a loophole for the local hidden theorist who wants to hold fast to both QM and to local hidden variables, then you may say that the denial of superdeterminism is an axiom of Bell's theorem. What else you consider to be an axiom would depend how you are formalizing Bell's argument (in support of his conclusion) and what commonly agreed presupposition(s) you are attempting to question.

But you are ducking my main point that the denial or affirmation of superdeterminism has very little bearing on the issue of the freedom of the human will when you consider that there are compatibilist accounts of free will, on the one hand, and also that quantum indeterminacies are commonly regarded not to provide the sort of leeway that ascription of human freedom and responsibility require, on the other hand. Human freedom from predetermination isn't the freedom to behave in accordance to the result of God's dice throws on almost all accounts (if you would except accounts such as Robert Kane's quite sophisticated QM dependent version of libertarianism).

Often the Free Will Axiom is called the "free will loophole", or the "free choice of detector orientations". — tom

You are misunderstanding Bell's statement. Bell's theorem is derived from the assumption of local hidden variables. Hence, if experiments show Bell's inequalities to be violate (as indeed Alain Aspect was the first to show experimentally) then it follows that either there a no hidden variables, or, if there are hidden variables, then measurement on one member of an entangled pair has a causal effect on the other element, and this causal effect is transmitted faster than light in violation of the special theory of relativity (and in violation of "local realism"). The assumption of superdeterminism -- that, somehow, the orientations of the measurement apparatuses as well as the measurement results all are predetermined in accordance with the statistical predictions of QM (an hypothesis rather akin to Leibniz's preestablished harmony) is one way to save local variables consistently with the predictions of QM, albeit a metaphysically extravagant and rather gratuitous way to do so.

Bell, like many physicists, also takes for granted that determinism and free will are incompatible and hence views the denial of determinism as being tantamount to the denial of human free will. He just assumes incompatibilism.

In summary, the "free will axiom" only is required provided "free will" is understood to mean "indeterminsm", which therefore excludes supercompatibilism. Since few physicists endorse superdeterminism (because they don't care for hidden variables or for local realism) they take QM to be an indeterministic theory, and, in this sense only, they are endorsing the "free will axiom" (a phrase that is almost never used in the literature).

This is nonsense. All of science implicitly assumes the free will of the experimenter. In QM this is made explicit in Bell's Theorem and various similar theorems. Otherwise we are super-determined — tom

You are running together the concepts of determinism and the concept of free will. Only if you assume the validity of a specific form of philosophical incompatibilism (i.e. the incompatibilism of free will with indeterminism at the level of physical law) are the two concepts coextensive. The only thing that is generally acknowledged by theoretical physicists in connection with the entanglement issue that you raise up is that the state of the measurement apparatus (e.g. whether is it set to measure some variable A or rather the conjugate variable B) and the ensuing measurement result, are not jointly predetermined -- i.e. there does not exist "hidden variables" which those states are pre-determined by and which we are merely empirically ignorant of. Those are the hidden albeit real "elements of reality" argued for by Einstein in the famous EPR paper. What the empirical verification of Bell's inequalities establishes is that Einstein was wrong about that. There are no hidden variables and an entangled electron (for instance) doesn't have a determinate momentum prior to a measurement having been effected either on this electron or on the other electron that it is entangled with. Hence, QM genuinely is indeterministic. Some physicists use the label "free will" to designate this lack of pre-determination. It merely amounts to a rejection of hidden variables. But it has nothing to do with the philosophical issue of the freedom of the will. Action, practical reason and personal responsibility are not concepts of theoretical physics at all.

And that is one of the reasons thermodynamics is *not* regarded as a fundamental theory. — tom

The idea of QM and GR being fundamental while thermodynamics would be merely contingent and derivative (e.g. dependent on a low entropy initial state of the universe) is a contentious proposal. Some physicists view thermodynamics to be fundamental (e.g. Fermi, Feynman or Penrose, if I remember). In any case, the proposition is more of a matter of the interpretation of physical theory than it is a scientific result. Even if it were merely contingently true in our actual universe that the laws of thermodynamics are valid (i.e. that they are valid in our branch of the "multiverse" that, as it happens, has a very low entropy early state) then there would still exist a definite arrow of time, however contingent, and the laws that govern the evolution of the observable/measurable states of matter would still be non-deterministic. Quantum indeterminacies would still rule the day. And this fact, again, would have very little bearing on the philosophical issue of incompatibilist/compatibilist regarding free will and responsibility since most philosophers regard the existence quantum indeterminacies to be irrelevant to the freedom of the will (excepting a few, such as the libertarian Robert Kane). If God throws dice to establish some of our brain states while we are making decisions, that doesn't makes us any freer than we would be if he would simply constrain the evolution of those states to obey deterministic laws.

I am an ardent advocate of science as a method and a body of work but against metaphysical naturalism, and I think the two things are confused in determinism/freewill debates. — mcdoodle

Naturalism has many guises, some of which are distinguished in the various essays collected in De Caro and Macarthur, Naturalism in Question, HUP, 2004. I am partial to the sorts of naturalism espoused by Hilary Putnam, John McDowell and Jennifer Hornsby (who each have an essay published in that volume). Can you elaborate a bit on what sort of naturalism you are labeling metaphysical naturalism?

and especially its last long paragraph is almost impossible to disentangle — The Great Whatever

If we can't disentangle it, then we may need to deconstruct it ;-)

So even under determinism, one can distinguish having a choice from having no choice. [...]
And what one can distinguish has meaning. My having a determined choice means that my choosing determines the event, and having no choice means that my choosing has no effect. — unenlightened

Quite so!

Second, I think 'free will' is an idea unrelated to determinism. Its history is theological and in contemporary debates it remains akin to theology, a way of relating a person's view of psychology to their view of ontology. — mcdoodle

It is a good thing, even when one is a naturalist, that one's philosophy of mind not conflict with one's metaphysics or with one's ontological understanding of living beings. One's desire to avoid such conflicts need not be a covert attempt to save supernatural belief.

This is what I don't get. Under determinism, what happens is a sensitive function of the initial conditions at the big bang, or if you prefer the conditions at any other time. Choice cannot exist, neither can "testability". Playing word-games to preserve moral responsibility seems utterly futile. — tom

Whenever you are deliberating about what to do, you are making a principled and pragmatically justified distinction between those options that you either lack the capacity or opportunity to do, on the one hand, and those options (W, X, Y, Z, etc.) that are genuinely open to you, on the other hand. When you then settle for one of those options -- to do W, say -- and proceed to do it, it doesn't reveal the other options -- X, Y, Z, etc. -- retrospectively not to have been really open to you. It's not just because you do not chose to do something that you are thereby shown not to have had the opportunity or capacity to do it at all. That you merely didn't chose to do any one of those things only reveals that you didn't have a good reason to do them, or that you didn't engage in practical deliberation as well as you should have, maybe. The fault would thus lie within yourself (in your rational or moral character, say) rather than in the external antecedent physical circumstances of your action (where those "circumstances" include your own antecedent neurophysiological states).

You may then appropriately kick yourself for not having done the right thing. If you would rather hold the universe causally responsible for your not having made the right choice, owing to the past physical conditions required for you to make the right choice not having been realized, it may reveal that you are philosophically confused. You forgot that you are a part of the empirical/material universe and of its unfolding rather than the past physical "state" of the universe being external to you. On the former view, you are an embodied rational animal endowed with real cognitive abilities, whereas on the latter view you are just a temporally unextended node in a chain of physical events. The latter view portrays you has having fewer powers than a simple sugar cube is commonly held to have. Physical theory doesn't force this metaphysically extravagant picture of human beings upon us.

Are you sure? Compatibilism seems more like "a person is to blame for their choices, even though 'choice' doesn't exist".

"Could have done otherwise" doesn't mean anything under determinism. If "could" refers to anything real, then determinism does not hold at that point - i.e. either the laws of physics are wrong, or our understanding of them. I don't think compatibilists complain too much about physics. — tom

The phrase "could have done otherwise" can point to a human ability. Abilities are similar to dispositions. Dispositions exist (i.e. they are actual properties of things) even when they are not actualized. A sugar cube is soluble because it would dissolve if it were immersed in water. Even if you don't immerse it in water, it remains soluble. You don't say that it was insoluble during the time when it was dry. Even when the sugar cube is dry, it retains the disposition to dissolve in water.

Likewise, you can say of Sue, who got to work late, that she could have arrived at work on time to mean that she had the capacity and opportunity. That the capacity wasn't actualized when she got late doesn't entail that the capacity itself wasn't there. If Sue gets late to work while having the capacity and opportunity, then she can be blamed (unless she had a good excuse to arrive late intentionally). Only if she didn't have the capacity (e.g. being paralyzed by a stroke) or didn't have the opportunity (e.g. because her car had been stolen), or both, do we normally say that she could not have done otherwise.

(The substantial other options for semantics as I read them are proof-theoretic semantics, i.e. inference as the basis of meaning, championed by Dummett...or to abandon the analytic approach and accept a form of Bakhtinian dialogism, i.e. all is dialogue and 'true' would be just one of many markers that interlocutors would have some sort of agreement or score-keeping about) — mcdoodle

I think truth-conditional semantics exhibit some of the flaws that you notice owing possibly to misguided attempts by their advocates to construe them a means to factor apart purely semantic representational functions from the pragmatic features of language. Truth-conditional theories of meaning need not be construed in this peculiar reductive analytic way, and, I think, some deflationary theories of truth such as the so called identity theory of truth espoused by John McDowell and Jennifer Hornsby, and relied on by Sebastian Rödl (in his Categories of the Temporal: An Inquiry into the Forms of the Finite Intellect) are consistent (as they are intended to be) with Davidson's original programme provided that it is interpreted in a deflationary manner, and also that it is not divorced from Davidson's mature conception of radical interpretation as providing the basis for assignments of meanings to terms of a language used by a community of speakers where the goal of the interpretation is to rationalise their behaviors and not merely to interpret the utterances that they blurt out passively in specific perceptual contexts. (Sorry for the long sentence!) This approach, which relies on the more mature Davidson (who has distanced himself more from Quine), I think, combines artfully Wittgensteinian pragmatism with the theoretical resources that formal semantic theories make available for displaying the generative/combinatorial structure of language from within its embedded functioning in the life of embodied agents.

It is no accident that the three authors mentioned above have worked extensively in the philosophy of action and have been much influenced by Aristotle, Wittgenstein and Anscombe.

(See also Michael Luntley, Contemporary Philosophy of Thought: Truth, World and Context, Blackwell, 1999, for a useful guide to such "embodied" and deflationary approaches to truth-conditional theories of meaning)

My point was that in being "a public demonstration", this means that even empirically "objective" is really "subjective", the only difference being that the agreement expressed is collective. — apokrisis

I think such pragmatist accounts often encounter much resistance owing to the widespread tendency to understand "subjective" and "objective" to express contrary notions. See also David Wiggins, A Sensible Subjectivism? reprinted in his Needs, Value, Truth: Essays in the Philosophy of Value, Third Edition OUP, 1998, for the argument that those two concepts register independent features of classes of human judgments.

Inasmuch as judgments of truth refer back to intersubjectively shared standards of assessment, and de facto grounds of agreement in human sensibility, they are subjective. Inasmuch as they issue from a fallible capacity to judge, and can be shown to be in error by those very same standards of assessment, they are objective.