Free Will?

According to the Copenhagen interpretation of quantum theory the outcome of physical processes cannot be predicted with certainty because they are not determined with certainty. By this interpretation nature does not dictate but rather allows a number of different eventualities; each with its own probability.

There are at least two responses to Copenhagen. The first and some would say naïve approach is to accept that nature is governed by stochastic processes. The second approach endorsed by Hawking_[1] is to accept a new form of veiled determinism that could be resurrected from the death of Laplace’s demon; ‘given the state of a system at some time, the laws of nature determine the probabilities of various futures and pasts rather than determining the future and past with certainty……[but]……. our use of probabilistic terms to describe the outcomes of events in everyday life is a reflection not of the intrinsic nature of the process but only of our ignorance of certain aspects of it’.

Although quantum theory cannot predict when a particular particle will decay suppose each particle has a hidden clock that specifies exactly when the particle will decay. If this were the case what we assign as probability arises not from a stochastic process but from our ignorance of a deterministic variable; namely these hidden clocks. This type of system constitutes a hidden variable interpretation which hopes to recover determinism from probabilistic quantum theory.

But this deadlock was resolved by Conway and Kochen; ‘The axioms SPIN, TWIN, MIN imply that the response of a spin 1 particle to a triple experiment is free – that is to say, is not a function of properties of that part of the universe that is earlier than this response with respect to any given inertial frame_[2].’

Conway and Kochen_[3] go on to conclude ‘although determinism may be formally shown to be consistent, there is no longer any evidence that supports it, in view of the fact classical physics has been superseded by quantum mechanics, a non-deterministic theory. The import of the free will theorem is that it is not only current quantum theory, but the world itself that is non-deterministic, so that no future theory can return us to the clockwork universe’.

But does the freedom of quantum particles necessarily mean persistent macroscopic systems will also be free? Neurobiological findings have led to their own two-fold conclusion; namely 1) my choice at any given moment is restricted by my particular brain-state within a particular set of environmental conditions and; 2) decision-making processes generally occur at a level beneath personal awareness.

John-Dylan Haynes for example identified fMRI patterns of neural activity several seconds before the subject became aware of their decision to press either a right or left button. Haynes_[4] concludes, ‘how can I call will mine if I don’t even know when it occurred and what it decided to do?’ 

Such instinctive behavior that lies beneath the level of consciousness is of course modus operandi for simple life forms. Martin Heisenberg describes how Escherichia coli move around by rotating their flagellum. In one direction the bacterium drives forward while the reverse direction tumbles the bacterium into a new random heading. Depending on sense data the bacterium will either drive forward toward the stimuli or tumble randomly to escape it. This behavior demonstrates how simple bacterium respond dynamically to sense data by modulating their ‘random walk’ in a manner that helps them find food and the right temperature. There is an interplay between a random walk and an underlying deterministic algorithm that chooses flagellum direction in an obvious Cartesian stimulus-response way.

But according to Heisenberg_[5] this mechanism does also finds more profound application in more complex systems; ‘as with the bacterium’s locomotion, the activation of behavioral modules is based on the interplay between chance and lawfulness in the brain. Insufficiently equipped, insufficiently informed and short of time, animals have to find a module that is adaptive. Their brains, in a kind of random walk, continuously pre-activates, discards and reconfigures their options, and evaluates their possible short-term and long-term consequences’.

Can such interplay ultimately underwrite non-reductive physicalism; or do all such accounts, once fully understood inevitably give way to bottom up causality?

According to Murphy_[6] ‘the components of systems include processes. Processes become what they are due to the way they are organized, related to other component processes. Systems are stable structures of processes that create their own components by constraining the degree of freedom of processes so as to coordinate them with other processes. The coordination requires not just matter and energy but also information….system thinking rejects the earlier bias in favor of concrete entities over processes; it recognizes that complex wholes can be more that aggregates. It employs the concepts of boundary conditions, structures, information, feedback, and downward causation.’

Murphy describes the principle with a simple thermostat. The thermodynamic processes on the atom level interact in a predictable way to deflect the bi-metal strip in response to atomic bottom up causality. The predictableness of the deflection is used to create a two state device where it is either ‘too hot’ or ‘too cold’. But what constitutes the state change? That information does not come from the atoms underlying the thermostat’s operation but rather from the wider system; say a settable knob. Thus even the simplest of states is subject to both bottom up causal influence and boundary conditions.   

In the same way homeostatic feedback that regulates our metabolic rate has built in states. These states are hard coded into our DNA in exactly the same way that the bacterium’s algorithm is hard coded in its DNA and the thermostat’s state is hardcoded by the knob. We cannot attribute the change of state in the thermostat to bottom up causes and equally we cannot attribute the homeostatic goals only to bottom up processes.

To fully explain the system we need to move from a reductionist approach toward a systems approach. Deacon_[7] defines first order or supervening emergent systems as non-recurrent systems in which lower order relational properties are the constitutive factor determining higher-order properties.

Second order or simple recurrent emergence occurs when there is symmetry breaking. This is commonly found in non-linear systems where higher order regularities become unstable. The unpredictability coined in the phrase chaos system derives from the fact that the regularities at lower levels are strongly affected by regularities emerging at higher levels of organization. These types of systems are acutely sensitive to initial conditions and often amplify perturbations to undermine prior states. This undermining of states in certain circumstances allows the system to move from pre-determined goals as found in homeostatic systems, thermostats and bacterium locomotion to adaptive goals that can be refined by near real time environmental influences.

Third order or recurrent-recurrent-trans-scale emergent systems involve some form of information storage on top of recurrent emergent states. History now exerts a cumulative influence over the causal future of the system. In other words where information is stored, constraints derived from specific past high-order states can be re-entered into the lower-order dynamics leading to a further altering of lower order states. The system is not only subject to change based on currently undermined and goal orientated feedback but also subject to change based on stored outcomes that either amplify or minimise lower states.

Third order emergence constitutes the origination of information, semiosis, and teleology in the world. Murphy_[8] describes the hierarchy this way; ‘third order (evolutionary) emergence contains second-order (self-organizing) emergence as a limiting case, which in turn contains first order (supervening) emergence as a limiting case. For this reason it is insufficient to describe mental phenomena as merely supervening on cellular-molecular interactions. The many levels of embedded evolutionary emergent processes characteristic of brains is what enables them so rapidly to selectively amplify such a vast range of possible forms of activity’.

In this arrangement causality flows both ways. But how can a system downwardly affect the very components that constitute itself? Juarrero_[9] suggests that it is not forces and particle interactions that empower downward causation but rather constraints. Constraints pertain to an object’s embededness within its environment and its relationship with other objects in a given entrained system.

Such relationships arise not only from atomic entities and their energetic interactions but also from information contained within the system. Information theory distinguishes between context-free and context sensitive constraints. A throw of a dice is context free since its probability is independent of previous throws. On the other hand drawing an ace from a pack of cards is context-sensitive since previously drawn aces shift the odds.

Higher level systems are influenced by context-sensitive outcomes that do not influence components in a forceful manner but rather the system is such that changes in components influence the probabilities of the occurrence of lower level events. Complex adaptive theory postulates such systems become causal players in their own right partly independent of the behaviour of their own components, selectively influenced by the environment and capable of pursuing their own goals.

The behaviour observed in Haynes experiments are consistent with an instinctive process dominated by bottom up causation. But in the case of the human brain, having instincts is not the full picture. Rather our brains are multi-tiered structure that enables both bottom-up instinctive processes that remains largely below the subconscious horizon as well as more complex conscious processes of which we are aware.

John Maunsell_[10] describes the same interplay within visionary systems; ‘what we actually perceive is not the image on the retina, but a neural image formed in the cortex….an image is not a completely accurate representation of what is going on in the world; it has been adjusted. That adjustment can occur either by so called top-down processes involving voluntary decisions – such as the choice to focus our attention on the search for a red book in the shelf, or a friends face in the crowd – or by bottom up influences over which we have no control’.

Murphy_[11] concludes, ‘if the causal capacities of complex entities were nothing but the combined causal effects of the entities’ constituents, and if the most basic constituents operated according to deterministic laws, then it would indeed seem to be the case that humans could do nothing other than what their atoms, in aggregate do…..we have argued that this picture is wrong on three counts. First, it is widely accepted that the atoms do not behave deterministically. Second, it is becoming more and more widely recognized that complex dynamical systems can exhibit new sorts of causal capacities not found at the level of their constituents. We have emphasized, among these, sentience, goal seeking, consciousness, acting for a reason, and self-evaluation. Third we have argued that higher level systems exert downward effects on their constituents via selection among possibilities generated randomly, probabilistically, or according to deterministic lower level laws’.    

In sum neurological data largely denies the libertarian free-will thesis but the dynamicity of quantum mechanics coupled with the sensitivity of non-linear systems equally resists hard determinism.

At play then is a tension between the two. This an important theological point for Green_[12] who cites the ‘virtual absence in Paul of the language of forgiveness of sins. Sin needs to be addressed, but given Paul’s perspective on sin as less act and more disposition or compulsion mere forgiveness is insufficient…..required rather is human change. A theological transformation – a deep-seated conversion in one’s conception of God and thus, in one’s commitments, attitudes, and every day practises. Not surprisingly then what Paul promises in Romans 5-6 is not remission of sin but liberation from our enslavement to sin and decay’.

According to Green what we see in modern neuroscience is consistent with Paul who taught not libertarian freedom but instead painted the following more sober portrait;

1. We do what we are. Our behaviours are generated out of, and so reflect, our character and dispositions.
2. Who we are is both formed and is continually being formed socio-culturally, and especially relationally.
3. Choice is contextually determined, especially vis-à-vis ongoing relational influence and self-reflective contemplation on the basis and futures of past and prospective decisions.

Green believes the scriptural case for libertarian freedom has been greatly exaggerated. Instead scripture affirms that as neurobiological beings we are slaves to our genetic disposition, social experience and environmental influence and as such any freedom attributed to our actions can only be affirmed through the transformation of the mind. 

This is succinctly captured in Paul's letter to the Romans - ‘do not conform any longer to the pattern of this world, but be transformed by the renewing of your mind. Then you will be able to test and approve what God's will is - his good, pleasing and perfect will (12:2)’.

---

[1] The Grand Design  Hawking S Mlodinow L 2010 Bantum Press

[2] SPIN Axiom: Measurement of the squared (components of) spin of a spin 1 particle in three orthogonal directions always gives the answers 1,0,1 in some order

MIN Axiom: Assume that the experiments performed by A and B are space-like separated. Then the experimenter B can freely choose any one of the 33 particular directions w, and a’s response is independent of this choice. Similarly and independently, A can freely choose any one of the 40 triples x,y,z, and b’s response is independent of that choice.

TWIN Axiom: For twinned spin 1 particles, suppose experimenter A performs a triple experiment of measuring the squared spin component of particle a in three orthogonal directions x,y,z, while experimenter B measures the twinned particle b in one direction, w. Then if w happens to be in the same direction as one of the x,y,z, experimenter B’s measurement will necessarily yield the same answer as the corresponding measurement by A.

[3] The Free Will Theorem Conway and Kochen 2006 Found. Phys 36 – 1441-1473  & Strong Free Will Theorem 2009

[4] Nature News Website ‘Neuroscience vs philosophy; taking aim at free will’ 2011 Kerri Smith

[5] Nature 14 May 2009 p165 Heisenberg M

[6] Did my neurons make me do it? Murphy N and Brown W 2007 Oxford Press

[7] Quoted from: Did my neurons make me do it? Murphy N and Brown W 2007 Oxford Press

[8] Did my neurons make me do it? Murphy N and Brown W 2007 Oxford Press

[9] Intentional Behaviour as a Complex System Juarrero A 1999

[10] Visual Systems Provides Clues to How the Brain Perceives Science 275 (1997) 1583-85

[11] Did my neurons make me do it? Murphy N and Brown W 2007 Oxford Press 

[12] Body, Soul and Human Life: The nature of humanity in the bible. Green J. B. 2008 Baker Academic