Can every important issue be scientifically investigated and, by comparing enough experiments with mathematical theories, answered conclusively? Can no philosophy uncover truths beyond the reach of science?
Here I argue that, despite the great service it renders humanity, certain biases, intrinsic to science, must be investigated from outside it. I will mention two examples where philosophical considerations can identify limits to the scientific method:
- Freewill
- Neurochemical Evaluations of Happiness
Why Free Will Is Scientifically Unprovable
An increasing number of scientists and psychologists reject the existence of free will. Indeed 86% of philosophers reject libertarian free will, with most believing that all our actions, down to the most minute decisions, are ultimately determined by circumstances outside our control (though compatibilists believe, so long as our desires align with our actions, we are acting freely even though we could neither act nor desire otherwise).
Is this slide towards determinism due to scientific discoveries… or is it the result of a bias in the scientific method itself?
Let us review what science is:
Science is a process whereby relationships of cause to effect are hypothesized and framed as competing theories. Some important predictions of these competing theories are then tested through experiment and observation. The theory with the greatest predictive power is then provisionally accepted while theories with lower predictive powers are usually rejected as false.
In other words, science is the inherent activity of deriving laws with ever-increasing predictive power. The game of science is the game of prediction. Yet prediction is only possible in a deterministic system.
Trying to use the scientific method to prove the existence of free will is like trying to use an earth excavator to fly to the moon. How can an intellectual activity that grants credibility to new hypotheses based on their predictive ability ever possibly grant credibility to a hypothesis that our behaviour is somewhat unpredictable due to our choices not being fully predetermined?
Science can only disprove free will and prove determinism – it is structurally incapable of the reverse. Science can only validate free will to the extent it fails to prove determinism – or runs up against its own limitations.
And yet…
…despite the inherent tendency of the scientific method to accept ever more deterministic descriptions of the physical world, and the unprovable nature of free will, it’s hard to see how our current scientific knowledge based upon experiments could be more favourable to libertarian free will.
The requirements of libertarian free will are quite philosophically demanding. If every action was exactly determined by preceding circumstances, then there would be no freedom, yet if every action was completely uncorrelated with subsequent actions, there would be no will. The notion of will is that an agent’s wishes determine some of its important future actions so as to materialize that wish in reality. These simultaneous requirements seem almost self-contradictory, yet a system whose behaviour is predictable – and heavily predetermined – in the short term, yet unpredictable in the long term, would satisfy the necessary criteria for a modest interpretation of libertarian free will.
Absolute free will (including the freedom to teleport yourself to Mars) is omnipotence. No advocate of libertarian free will, or any sane human being, would seriously claim we possess this. Libertarian free will is simply the modest claim we possess some free will; that our brains can make some important conscious decisions; that some conscious decisions, which arise from our minds, produce physical actions with physical outcomes that were not always predetermined; that we could have decided differently and – had we done so – a different physical action and outcome could have arose as a result of that different decision (that if you wind back the clock far enough to the same past condition and run reality again – a different outcome could result).
The uncertainty principle, states that the position and momentum of particles is fundamentally undetermined. The Heisenberg uncertainty principle is more than just a limitation in our ability to measure particle position, as theses diagrams of electron orbitals in the hydrogen atom show. Each orbital represents the spatial probability of finding a single electron in a particular location in the atom. The p-orbital, for example, contains two dumb-bells of finite probability separated by an infinite plane of zero probability. The zero probability plane completely separates the finite probability regions from each other. There is no possible path for the electron to get from one finite probability region to the other without crossing the zero probability plane, yet if the electron didn’t cross the zero probability plane, the electric field and chemical characteristics of the hydrogen atom would be different. Thus, the electron must simultaneously exist on both sides of the zero probability plane.
Some theories appeal to an infinite number of unprovable parallel universes to interpret wave function collapse as being deterministic, but if we go back to the core value of science: experimental data is king. And experimentally, wave function collapse is non-deterministic.
While the short-term wiggle room for free will is limited to one atomic radius, the butterfly effect means very small differences in initial conditions can give rise to large differences in final outcomes. Here is the first observed example of a computer model exhibiting divergent trajectories in phase space.
The initial conditions vary by less than one part in a million, and, initially, the evolution of the system in both cases seems identical, but, after a period, the trajectory of both runs start to diverge and, eventually, they exhibit some important yet completely different behaviours.
The combined scientific observation of fundamental quantum uncertainty and long-term divergent behaviour from similar initial condition is conducive with short term macroscopic determinism and long-term macroscopic indeterminism. In other words, quantum uncertainty, when combined with the Butterfly Effect, leaves room for free will in psychology.
It is hard to imagine a physical description more favourable for the philosophically demanding requirements of libertarian free will than those routinely used to describe the behaviour of complex natural systems.
However, this may indicate free will has a long actuation time, which the Libet experiment cannot refute. Total short term freedom would, after all, negate the existence of a meaningful will.
Therefore, scientists who reject free will do not do so due to any scientific discovery (indeed discoveries in physics could scarcely infer a universe more favourable to free will) but rather due to of biases innate to the scientific method itself.
To complain there is no causal mechanism for free will and, therefore, that it doesn’t exist is to completely miss the point. Causal mechanisms are, by nature, deterministic. A causal mechanism for free will is an oxymoron and demonstrates how the tool box of science is fundamentally unequipped to prove the existence of free will. If no scientific experiment could ever falsify determinism, then claims for determinism are as unscientific as claims for free will.
Neurochemical Evaluations of Happiness
Given the importance we place on happiness, it’s worth asking: “What exactly is happiness?”
Understandably, many neuroscientists and psychologists have brought their expertise to bear on research in this area. Scientific enquiries into conditions for producing happiness are both valuable and important. But is there any hard limit to the authority and expertise that progressive experimental discoveries could convey to researchers studying happiness (and other subjective feelings like misery)?
Consider this thought experiment:
Jim walks into an MRI scanner where his brain activity is thoroughly analysed. After emerging from the machine, the researchers, who have decades of experience studying neurochemistry, say to Jim: “We’ve analysed your brain chemistry and the results are conclusive. You are acutely depressed.”
Jim responds: “But I feel happy! I rarely get depressed.”
To which the researchers respond: “You are clearly mistakenly evaluating your own experience. The results of the MRI are conclusive. You are acutely depressed. You must register for a happiness enhancement course. It’s for your own good.”
Jim responds: “But I don’t need a happiness enhancement course, I already am happy.”
To which one researcher responds: “Jim, I have a 1st class degree in neuroscience, a Phd. in evaluating the neurochemistry of happiness and depression. I have spent the last 15 years looking at different people’s brain activities and evaluating them for happiness, or misery. The other 10 researchers on my team all have similar levels of experience and, after analysing your brain activity, we have all reached the same conclusion: that you are thoroughly depressed. How much experience do you have in analysing brain patterns?”
To which Jim responds: “None.”
The researcher then asks: “So why do you think you’d have more knowledge about your brain patterns than a team of researchers who’ve spent decades of their careers measuring and analysing brain patterns to decide whether patients are happy or miserable?”
Question:
Could any amount of scientific advances, cranial measuring equipment or AI data analysis techniques in an infinitely sophisticated future ever conceivably make it possible for the team of neuroscientists to be right and for Jim to be wrong?
I take the view that, for any sensible definition of happiness or misery, the answer to this question is “no.” All happiness evaluations must begin with a questionnaire. Now through comparing questionnaires with measured brain activity, facial expressions or other kinds of behaviour, we may find reliable markers that correlate well with reported happiness (dopamine levels, serotonin levels, activities in particular brain centres etc.). In the future, neuroscientists may, by monitoring brain activity, be able to anticipate reported happiness with 99%, 99.99% or 99.99999% accuracy. So, if neuroscience can achieve such accuracy in predicting how someone reports feeling, what happens when that one person reports a very different feeling to what the instruments tell the scientists he should feel? Are the scientists wrong about how the subject feels – or is the subject wrong about his own feelings?
Given that reported feelings are the foundation for correlating brain activity to subjective experience, subjective present experience must always have absolute sovereignty.
One could, of course, scientifically define happiness in terms of dopamine levels much like we define force as mass times acceleration, but such a definition would be irrelevant to the inherent philosophical value of happiness as a subjective sense of satisfaction, joy, pleasure, freedom from suffering, etc., etc., And there probably is a danger that, in the future, overzealous neuroscientists might favour scientific definitions of emotions like happiness, anger, sadness, depression that are easier to objectively measure and stick into computer models, even if such definitions are less relevant to the subjective emotions that we feel and value. There’s also a tendency in some fields to ignore outliers, in order to get journal papers published. All of this, if unscrutinised, could result in experts overriding people’s subjective experience.
But:
It is certainly possible that, in the future, neuroscientists will gain sufficient understanding of the workings of the brain to anticipate your future subjective experience, or evaluate your past subjective experiences, better than you can yourself.
The absolute sovereignty of the individual in evaluating their own experience, only applies to their immediate present experience. Expertise can (and probably someday will) anticipate people’s future subjective experience with greater accuracy than they themselves can. And there is already some important evidence that we aren’t good at objectively evaluating (or accurately remembering) how happy we were during an extended past period. People’s evaluation of their level of happiness over their entire lifespan varied noticeably based on whether they had recently found a few dimes left by a photocopying machine, suggesting our present mood strongly influences how we evaluate past emotions.
Experts may also understand why a subject reports an experience better than the subject’s own belief for the reasons of his experience. Numerous experiments clearly demonstrate that people can frequently give erroneous accounts of the reasons for their motivations and, perhaps, even feelings.
But this does not alter our absolute knowledge of our immediate subjective experience itself.
But being able to predict what people will subjectively feel themselves in the future, or why they feel what they feel, better than they can themselves, is distinct from being able to dictate what people feel in the present, better than they can themselves.
There may also be a one-to-many problem that runs both ways, where many different brain-states in different people lead to similar subjective experiences, while nearly identical brain-states in different people may occasionally give rise to significantly different subjective experiences.
So the evaluation of people’s present experience through evaluating neurological activity is an example of where science can only catch up with, but never surpass, common personal experience.
John
Do You Have a Burning Desire To Make a Comment?
Have you found this article thought provoking? Is there some message you desperately want to communicate to future readers but can’t because my comment section automatically closes 28 days after my posts go live?
If so, you might be interested to know that I reopen any comments section to members of my mailing on request as one of the perks of joining.
If you’d like to leave a comment, simply scroll to the bottom of the page, sign on to my mailing list and them email me with a request to reopen the comments section for this post.
Happy Commenting!
John
That was a very deep analysis beyond the common understandings of the label of “science” which tends to be the new God. The levels of its authority seem huge which unavoidably gets it closer to the characteristics of a dogma. However, science by its definition is meant to produce everchanging findings the way I understand it.
So your two points were really eyeopening. First and foremost is the issue of the conditions of a system. The system science is not let by its conditions themselves to consider free will, that’s right! But in my opinion, there is more to that. The experimental methods are valuable and humanity needs them. What we need to change is the way we perceive them. Instead of taking a finding as a fact, we can be taking it as something very possible but also being open that it may not explain the truth.
And you explained it very well when you talked about the difference between an individual’s experience and several external observations of it. Behind that to be honest I recognize a tendency to undermine the individual and at the same time overestimate the authority.
Lastly, promoting a deterministic perception of reality in the very end underestimates humanity itself, forgets creativity and above all takes away any idea about consciousness. Because with no free will there are no choices. And with no choices there is no consciousness.
The good thing is that non determinism, as you wrote, is a more possible phenomenon even according to physics.
Yes, I too consider the experimental method to be very valuable and it can answer a lot. Including very important practical things such as the best ways to cure disease or the cause for spreading them. But we need to know the limits intrinsic to science in order to be use it properly.
…and it stands to reason that the limits of science may well be scientifically unprovable. Although the Heisenberg uncertainty principle is an example of a precise limit to science that science can prove and experiments can verify.
The notion of science always involves an observer looking at the behaviour of a system (i.e. something) from the outside. And while we now know that the observer can perturb the system by looking at it, this still cannot encapsulate the subjective conscious experience of the system.
We can only even guess at the subjective experience of another conscious entity through the introspection of our own conscious experience and then (in however flawed a manner) try to extrapolate that experience onto the other, perhaps asking questions and communicating with the other as well.
There seems to be a lack of introspection in science and this may limit its ability to grasp consciousness.
As the fellow at RationalRealm suggests, we never experience freedom as such. We experience a constraint, and when that constraint is released, we feel free (of the constraint). To be meaningful, the terms “free” and “freedom” must explicitly or implicitly refer to some meaningful and relevant constraint.
No one experiences reliable cause and effect itself as a constraint. Every freedom we have, to do anything at all, requires reliable causation. We, ourselves, are a collaborative collection of deterministic mechanisms , mechanisms keep our heart beating and our blood flowing. Mechanisms that input sensory data, organize it into a model of reality, enabling us to control many aspects of our physical and social environment.
To view reliable causation itself as some kind of boogeyman that robs us of this control is a perverse interpretation of the empirical evidence. We get seduced into such a view by a hoax, one that implants the suggestion that we are not “truly” free unless we are also free of causation. But “freedom from reliable cause and effect” is an irrational concept, because, without reliable causation, we could never reliably cause any effect, and would have no freedom to do anything at all. So, no use of the term “free” or “freedom” can possibly be taken to imply “freedom from causation”, including the “free” in free will. Because it cannot, it does not.
Free will is what we call it when a person decides for themselves what they will do, free of coercion and other forms of undue influence (hypnosis, mental illness, etc.). It requires nothing supernatural. It makes no claims against causation. And yet it is quite sufficient for both moral and legal responsibility.
Determinism asserts only that every state and event is reliably brought about by prior states and events through a specific series of reliable causes and effects. It cannot, and therefore does not, assert that something other than us is the meaningful and relevant cause of our deliberate choices and our actions, except when we are subject to coercion or other undue influence.
By the way, I do have an article that explores a number of deceptions that are used to create and sustain the paradox at https://marvinedwards.me/2019/03/08/free-will-whats-wrong-and-how-to-fix-it/ if you’re interested.
I’m inclined to agree with your view that our identity could not meaningfully exist in the absence of any causation.
There is, however, a distinction between one-to-one causation and one-to-many causation. Between a unique past giving rise to a uniquely defined future and a unique past potentially giving rise to a basket of alternative possible futures.
Furthermore, our current understanding of the physical universe, does not imply one-to-one causation. At the quantum level there is a high level of randomness that sometimes statistically smooths out to give deterministic one-to-one cause-to-effect outcomes (such as the moon orbiting the Earth or the motion of a bowling ball released) and sometimes may not (I’m actually not sure whether we know if out brains are systems in which quantum fluctuations may give rise to different outcomes or more computer-like systems with charge-packets large enough to negate quantum fluctuations).
“We ourselves are a collaborative collection of deterministic mechanisms”
That hasn’t been established. Chaotic systems have a tendency to oscillate around “strange attractors” where the systems behaviour is constrained within a range but has some freedom to unpredictably vary within that range. Obviously the mechanisms that power living systems must reliably function within a range conductive to promoting life (most of the time – until to get a stroke or something) but it certainly possible that our bodies how some wiggle room within that range.
I can agree with you about coercion, but I’m not sure I can agree that if our behaviour is fully determined by the past back until the moment of our birth that “undue influence” can be distinguished from “due influence.” I can see your point about the distinction between the driver of the getaway car thinking “I don’t want to help these scumbags…but I don’t want to die either.” But if our actions are fully determined by past events, there can be no meaningful distinction between someone who was hypnotized to want to kill another person and someone who wanted to kill another person for any other reason. In both case, each participant acted in accordance with their will and that will was fully determined by prior events (assuming the universe is fully deterministic).
I suppose the compatibilist idea of free will being the uninhibited realization of a mental desire is a consistent and defensible one. But there’s no getting away from the fact that if the universe is completely determined we have to face the fact that choice is an illusion.
Determinism does have big implications – there’s no avoiding it. We like the idea that we have the ability to make choices in life. The idea of installing microchips into people brains which make people want to do things in a remotely controlled way that was exactly pre-specified by someone operating a control panel behind the scenes would revolt many people – even if, by installing such microchips, people would lives happier lives.
Yet if the universe is fully deterministic, our existing choice is an illusion (even if some interpretation of “free will” might still be tenable).
So it’s not the case that determinism is a distraction of no consequence or a storm in a tea cup. It may be possible that some aspects of our existing justice system could still be defensible in a deterministic universe – but I very much doubt that all of it could be.
“But if our actions are fully determined by past events, there can be no meaningful distinction between someone who was hypnotized to want to kill another person and someone who wanted to kill another person for any other reason.”
And that is the problem. Universal causal inevitability makes no meaningful distinction between any two events. But free will, defined as “free of coercion and undue influence”, makes a meaningful empirical distinction between someone acting deliberately versus someone acting under the influence of hypnosis. This is meaningful because we must decide whether we arrest the person himself or we arrest the hypnotist.
Deterministic inevitability, being true in all cases, makes itself irrelevant by its own ubiquity. It is like a constant appearing on both sides of every equation, it can be subtracted from both sides without affecting the result.
“In both case, each participant acted in accordance with their will and that will was fully determined by prior events (assuming the universe is fully deterministic).”
But we cannot say causes prior to the person determined the event, because the most significant prior causes took place within a person. Either it was the prior deliberation of the person himself that causally determined the choice, or it was the prior deliberation of the hypnotist that determined the choice. And, unlike deterministic inevitability, the person’s behavior is something we can actually do something about, which makes it a relevant and meaningful cause.
I think that both “random” and “chaotic” refer to problems in prediction rather than problems in causation. Random events are deterministic, but difficult to predict, like flipping a coin. Chaotic events are also deterministic, but deviate quickly from predictions due to an inability to precisely control the starting conditions.
I suspect that quantum events will eventually turn out to be problems of prediction as well. Their rules may be different from the “laws” of physics, but we might hope that they are following their own rules which only apply at that level. We have a similar situation going up from physics to living organisms, and then intelligent species. Physics cannot explain why a car stops at a red light. It is not that any laws of physics are broken, it is just that they don’t cover everything. For example, we won’t find the laws of traffic in any textbook on physics. Nor will we find the driver’s motivation to survive and the rational calculation that stopping at the light will accomplish that goal.
How could you distinguish “due influence” from “undue influence”?
“Either it was the prior deliberation of the person himself that causally determined the choice, or it was the prior deliberation of the hypnotist that determined the choice.”
You’re implicitly appealing to a “first cause” which is the whole problem with determinism which asserts there is no first cause. What cause the person to deliberate? What cause the hypnotist to deliberate? Is the hypnotist any more a “first cause” than the person who was hypnotized? If both the actions of the hypnotist and the one who was hypnotized are perfectly determined by prior events and if they both acted according to their will (which in one case was implanted by hypnotic suggestion) then why not punish both?
“Deterministic inevitability, being true in all cases, makes itself irrelevant by its own ubiquity. ”
If we accept determinism we have a serious issue discriminating due influence from undue influence. Discriminating hypnosis and brainwashing from spontaneous choice (according to determinism spontaneity does not exist). You mentioned deliberation. What if the hypnotist hypnotizes his victim to deliberate.
If you think about it carefully, you’ll find these problems don’t just conveniently disappear in the way that you suggest.
“Random events are deterministic.”
Wrong.
Random events are not deterministic. They are, in fact, the absolute opposite of being deterministic. An event that is deterministic may be mistaken by an observer as being random. But that is simply the observer making a mistake, like point to a bat at dusk and calling it a bird. An event that appears random might be deterministic, but an event that is truly random is non-deterministic by definition.
“I suspect that quantum events will eventually turn out to be problems of prediction as well.”
If you paid attention to the article, you would see that the probability cloud of a single electron is split in half by a zero probability plane. This means the electron must be in two places at once at a factual level. This is not just a limit to observation.
The uncertainty principle also predicts the spectral broadening of lines in some instances where the excitation period is low, the uncertainty in the excited energy and the variance in wavelength of the emitted lines.
Hawking radiation is another phenomena that results from factual spontaneous oscillation in quantum energy.
https://www.technologyreview.com/s/420940/first-observation-of-hawking-radiation/
Indeed recent measurements of the cosmic background radiation confirm that random quantum fluctuations are what seeded the collapse of the first galaxies.
https://dailygalaxy.com/2015/06/quantum-origins-of-the-universe-seeded-the-early-galaxies-and-clusters-weekend-feature/
Are galaxies real enough?