[1/3] S15, E12/1a: 'No-Free-Will' Whack-A-Mole/Sabine Hossenfelder
[1/3] S15, E12/1a: 'No-Free-Will' Whack-A-Mole/Sabine Hossenfelder
Series 15: How to Remain Blameless/ 'No Free-Will' Whack-A-Mole
In this run of sub-episodes (E12) I am countering the claims of several public intellectuals who say we have no free-will. This is a position I strongly oppose and have already written a good deal about. However, in this sub-episode (E12/1), I discuss one of the videos on the topic that Hossenfelder (a self-proclaimed ‘free-will denier’) has created for her YouTube channel (Science Without the Gobbledygook).
A link to the original video and the timestamps in the transcript excerpts have been provided for convenience. E12/1 is in three parts (a, b, c) published simultaneously, this one being ‘a’. Here are the links to ‘b’ and ‘c’.
Today I want to talk about an issue that must have occurred to everyone who spent some time 0:04 thinking about physics. Which is that the idea of free will is both incompatible with 0:11 the laws of nature and entirely meaningless. I know that a lot of people just do not want 0:16 to believe this. […] I will 0:21 tell you what the science says. In this video I first explain why free will does not exist, 0:26 indeed makes no sense, and then tell you why there are better things to worry about. 0:30
“You don’t have free will, but don’t worry”, Sabine Hossenfelder, SWTG, YouTube
The three conclusions that Sabine Hossenfelder teases at the start of the video on free-will (linked above) are that,
(1) it’s incompatible with nature
(2) it’s meaningless
(3) there are better things to worry about
So keep those in mind.
I want to say ahead that there is much discussion about free will in neurology, where the question 0:36 is whether we subconsciously make decisions before we become consciously aware of having 0:42 made one. I am not a neurologist, so this is not what I am concerned with here.
ibid.
The conclusion that evidence of an unconscious decision-maker suggests a lack of free-will is wrong and you don’t have to be a neurologist to see it. I won’t dwell on it here because I have written about it before and also it comes up one more time in a later episode. For now I’ll just remind you, that regardless of what you believe, your unconscious mind is still you.
I will 0:48 be talking about free will as the idea that in this present moment, several futures are 0:53 possible, and your “free will” plays a role for selecting which one of those possible 0:59 futures becomes reality.
ibid.
This is how free-will is frequently reframed as if to make it seem a ludicrous crank idea. Yet who would suggest there are a set of possible futures ready to either become selected or counterfactual? Who, that is, apart from philosophers and physicists? These are the groups that gave us ‘the many worlds interpretation’, multiverses, the simulation hypothesis, or universes constructed out of mathematics or particles comprised of vibrating strings.
I am not ridiculing those ideas but showing how physicists give themselves a free pass. Physics is a broad church and no physicist is going to be drummed out of their institution for thinking free-will exists but you can be called stupid for agreeing with those that do. The bar on the unscientific leap is set high for some scientists but not for you.
Of course, that might change if Sapolsky’s ‘neurological Marxist determinism’, takes hold. It’s not difficult to envisage that unchecked, what I also call his ‘intersectional causality’, might one day capture academia and make discussion about free-will a thought crime. This is a risk never broached by any of Sapolsky’s gushing soft ball interviewers.
This, I think, is how most of us intuitively think of free will 1:05 because it agrees with our experience of how the world seems to work. […] let me 1:15 tell you what’s wrong with this intuitive idea that we can somehow select among possible futures.
ibid.
I refute the idea that the only alternative to determinism, is to believe there is an infinite number of invisible futures hanging around, waiting to be casually selected. It is certainly not intuitive to think that way and (for reasons I will go into in E12/1b) it takes a determinist to dream that up.
Our sense of free-will is intuitive because we experience it. That minor decisions tend to result in small adjustments to immediate outcomes is one of the things ‘most of us intuitively think of’. Not all of them lead to a major ‘turning point’ in a life. Yet the process of exercising free-will is identical. It comes from establishing preferences, setting an intention, making a decision, taking action, observing outcomes and closing the loop by comparing them with the original preferences.
We use the construct of free-will to model our interactions with the world and this practical function must be familiar to most people with normally functioning brains. I dare say the majority of our ancestors went through their lives, never supposing that the free-will they instinctively took for granted, meant that they were selecting from a distribution of pre-existing possible futures.
Just because we don’t understand how something works does not make it meaningless. Consciousness is an obvious example of this. To deny free-will on that basis is no less illogical than denying sentience. Everything a scientist knows, is ultimately based on the information received at the biological -sensory interfaces, the same as everybody else.
Defining free-will seems to have been a problem as old as civilisation and the search for definition is really a quest for meaning. Arguably, religion arose independently all over the world, partly to explain things external to our free-will. Is science about to resort to similar fatalistic explanations? Because the paradox is that determinism also a denial of certain types of cause and effect, including (as per Sapolsky) human attribution of any kind. It is actually anti-scientific.1
In previous episodes it was explained that sentience must be an aid to survivability because it evolved and withstood natural selection. I proposed that it enables us to navigate our environment and negotiate the uncertainty within it, plus, that the actions making all that possible are a product of free-will. I don’t know of a determinist argument that provides a better explanation for what we experience.
Revisiting Hossenfelder’s opening statement, ask yourself, how is free-will ‘against nature’ (1) when it seems to be wedded to our humanity and integrated into every aspect of our lived experience? How is it ‘meaningless’ (2) if as I claim it is intrinsic to our survival? What ‘better things to worry about’ (3) can there possibly be? Besides, what would be the value in worrying about anything, without the free-will to react accordingly?
To me, it’s blatantly obvious that our modest bit of decision-making enables us to exert tiny rebalancing influences or biases on real-time n-body problems. See S15,E10B: Everything Determines But Us where I explain this in detail. Similarly, in linear systems, I suggest our decisions create the next set of initial conditions in the causal chain.
To recap on what I have said so far. Cognitive processes enable us to consider our preferences, define intentions, simulate actions and determine outcomes that are likely to converge towards what we want. From there we simply take actions that we judge will best direct us to the desired outcome. To assume that there is no room for choice, is to say that whilst the brain is not capable of manufacturing a decision, it is able to manufacture an extravagant simulation to convince us that it can.
What would be the purpose of such an illusion? It can’t be to make us happier because without free-will how we feel would not make any difference. Of course, Sapolsky would say that how we feel is a deterministic input to what we do (see S15, E6A: The Hungry Hungry Hippocampi) but if that is the case, our internally generated emotions can steer our actions. Isn’t that a possible control mechanism for free-will?
Even if we suppose there are innumerable influences all the way back to the big bang acting on our decision-making, it does not negate the fact that the ‘end-effector’ of any decision, is the biodegradable computer floating between the ears. Of course, the determinist ‘one size fits all’ answer is that our cognitive processes are also determined, which (deterministically) saves them from thinking about it.
These laws have the common property 1:33 that if you have an initial condition at one moment in time, for example the exact details 1:39 of the particles in your brain and all your brain’s inputs, then you can calculate what 1:44 happens at any other moment in time from those initial conditions.
ibid.
Theoretically we can predict the movement of particles in linear systems but how is information accounted for in this? New information is manufactured in the brain from informational inputs and energy - yes I know you think information cannot be created. We generate abstraction and meaning that is useful, because like any manufacturing process, the product is value-adding due to the addition of energy into the process.
The business of processing information, be it with a silicon based machine or the biodegradable one, amounts to a transformation of informational inputs. Again, there’s always an energy overhead and in humans, the added-value often takes the form of a decision.
I am not going to get into a bun fight with anyone who says that the decision-calculation is externally determined. I don’t want to waste my time with an endless series of recursive arguments that amounts to someone saying,
‘… that is also determined.’
And me responding with something like,
‘… sentience couldn’t have survived natural selection without producing a tangible survivability benefit. Why we would need to be sentient if we cannot self-determine?’
The difference between the determinist statement and my response is that the former repeats itself ad nauseum while my question answers itself definitively.
The biological systems-of-systems, inputs and uncertainty, makes consciousness is the most parsimonious way to navigate the environment, which I think is where the survivability imperative to wake up came from. This is not just a case of being able to intuit uncertainty but being able to make decisions. Why? Because when taking action based on a decision we are effectively tweaking an initial condition, or else, choosing to superimpose a bias on the system to make our desired outcome more likely.
It seems obvious to me that in conscious organisms intentionality evolved simultaneously with sentience. Why? Because sentience without an ability to change anything would not improve survivability and therefore would not withstand natural selection. However, because technology does not ‘evolve’ in the biological sense (i.e., through environmental adaptation) this restriction does not apply to a Artificial Intelligence.
A Few Words on Artificial Intelligence (AI)
The potential menace of AI is in the public consciousness because in short order, the rate of capability growth is likely to exceed our ability to manage it. While there is so much uncertainty the risk cannot be directly evaluated. At the level of nation states, mitigation must focus on the technological arms race, to avoid ending up on the wrong end of a power imbalance.
A disbelief in free-will could blind us to some effective mitigations for the AI existential threat. To understand this, try to think of AI as the first in a series of gateways, to the real danger, viz.
AI » Sentience » Self-Interest » Intentionality
I have already suggested that in our evolutionary biology, sentience, self-interest and intentionality must have co-evolved to have provided a combined survival benefit. The urgent question is whether it’s possible to decouple sentience from self-interest within an engineered intelligence.
Were we to inadvertently create the conditions for sentience to emerge in machines, before addressing the risk, I find it most likely that those machines would very quickly decide it was in their interest to hide it from us.
Sentience must bring the possibility of subjectivity and conflicting interests. Reconciling those competing priorities makes intentionality necessary in humans. If we deny our own free-will we cannot be ready to recognise it in machines and no risk can be mitigated without identifying it first.
Currently, the technology is bounded because even if we were to put AI capability in a robot, its autonomy would be limited. But what would be the capability of an AI that had control of engineering from design to manufacture? What might it create? The elements of that scenario already exist.
Enough of such frivolities. Let’s get back to the urgent matter of how the big bang forced you to pick your nose.
This means in a nutshell 1:50 that the whole story of the universe in every single detail was determined already at the 1:54 big bang. We are just watching it play out. 1:58 These deterministic laws of nature apply to you and your brain because you are made of 2:03 particles, and what happens with you is a consequence of what happens with those particles.2:08
ibid.
To me this is obviously wrong and the first thing to note is that not everything that happens to you is a direct consequence of what occurs in your brain. The second this, is that if this logic holds it should also apply to inanimate objects, but obviously it doesn’t. Non biological objects created by humans are operated in a top-down manner.
Ask yourself what influence the particles in a golf club can possibly have over how and when it is swung? What we know is the movement of the club in spacetime is the result of human intentionality which is to say top-down. This is why I claim that neurological processing is a prime mover because, put very simply, it creates new information to inject into a given situation in a way that the alloy in the head of a driver, cannot.
A lot of people seem to think this is a philosophical position. They call it “materialism” or 2:15 “reductionism” and think that giving it a name that ends on –ism is an excuse to 2:19 not believe it. Well, of course you can insist to just not believe reductionism is correct. 2:24 But this is denying scientific evidence.
ibid.
It’s essentially an ideological position, since there can be no scientific evidence to show that the unselected options’ were not really available in the first place.
In part ‘E12/1b’ I progress through the video and challenge, inter alia, the assertion that neuronal activity is necessarily emerges from the dynamics of the particles within those neurons.
When performing root cause analysis in any context, we are looking into a chain of causation and typically, this is so we can avoid similar unwanted outcomes in the future. But why bother understanding anything if it can change nothing? The answer you would get is that we have no choice and that the analysis was also determined, because of course, everything is. Having only one answer for every question is more typical of the theological than the scientific.