Einstein on Free Will

by Max Andrews

After the First World War Einstein made contributions to the development of quantum theory, including Bose-Einstein statistics and the basics of stimulated emission of radiation from atoms (which was later used to develop lasers).  He gave the nod of approval that led to the rapid acceptance of Louis de Broglie’s ideas about matter waves but he never came to terms with the Copenhagen interpretation of quantum mechanics.[1] The Copenhagen has become the more popular and standard interpretation.[2]

According to the Heisenberg Principle, the moment at which a measurement takes place is the moment at which the randomness lying at the heart of quantum reality expresses itself.[3]  Up to that point, everything is fine.  Amplitudes change in a completely predictable, and more importantly, calculable way.  The observer changes the state of what is being observed.  Outcomes can be predicted according to governing probabilities, but the actual outcome cannot be known in advance.[4]

This was something Einstein could not live with.  Einstein, as a determinist, felt that the world is a structured and rigid web where effects follows cause and all things should be predictable, given the right information.  Einstein acknowledged that quantum theory works but he did not like the philosophy behind it.  If whether or not, for example, Niels Bohr, Einstein’s quantum physics counterpart, were to throw a book across the room Einstein would be able to predict the outcome of Bohr’s “choice.” Einstein would of course say that choice is the wrong word to use; rather, the brain is a complex machine with cogs whirring round to produce a predictable action.  The basis of Einstein’s view was a philosophical conviction that the world did not include random events:  an objection summed up in Einstein’s widely quoted saying, “God does not play dice.”[5]  Bohr is reported to have responded to Einstein with the witty reply, “Don’t tell God what to do.”

Strict [or hard] determinism may be the only way to avoid the implication from quantum mechanics and experiments such as the delayed choice experiment.[6]  This experiment suggests that quantum communications occur instantaneously across any distance, or even travel backwards in time.[7]  The determinist is not yet defeated, quantum mechanics comes with a state of collapse and that seems to be linked to measurement.  Whatever measurements are, they are very specific situations and probably linked to what happens when a particle bumps into a measuring device.[8]

Einstein played a prominent role in the early development of quantum mechanics, particular in his philosophical approach to it.  How one interprets quantum mechanics will shape the answer to the question of determinism and free will.  Empirical testing does not seem to be enough to provide a satisfactory answer; rather, it how the data is interpreted.  Einstein’s approach to the rejection of genuine random events has been an influence of the contemporary debate.  It has been argued that Einstein’s determinism is correct, but it may be a mistake for him to base it on random events.  Randomness is not sufficient for determinism to be true; a lack of causality would be sufficient.  Even with the delayed choice experiment there seems to be a lack of causality, if anything it would be backwards causality.  The free will proponent must be careful not to appeal to any ignorance for a lack of explanation of such quantum events.  Einstein’s reason for determinism (randomness) does nothing to advance his case.  If anything, quantum experiments such as the delayed choice experiment only show that there is randomness in the world, not that there is purposeful, free agency.  All quantum mechanics entails is that there are random events in the brain (or whatever) that yield unpredictable behavior, which the agent is not responsible.[9]  Thus, it seems to be the case that Einstein’s philosophy of determinism has persevered.[10]


[1] Kenneth William Ford, The Quantum World: Quantum Physics for Everyone (Cambridge, MA: Harvard University Press, 2004), 117.

[2] At this time there are at least ten regularly cited interpretations of quantum physics varying in interpretation of wave collapse, determinacy/indeterminacy, superpositions, and Schrödinger’s equations.

[3]  The equation: (change in x multiplied by the change in px is greater than or equal to half of Planck’s constant). For a given state, the smaller the range of probable x values involved in a position expansion, the larger the range of probable px values involved in a momentum expansion, and vice versa.  The key to the expression is the greater than or equal to because it places a limit on how precise the two measurements can be.  The principle is relating and for the same state ( signifies change, h, h-bar, is the Planck constant).  Heisenberg’s target was causality. The Copenhagen interpretation adopted this principle.  Jonathan Allday, Quantum Reality: Theory and Philosophy (Boca Raton, FL: CRC Press, 2009), 247-248.

[4] Jonathan Allday, Quantum Reality: Theory and Philosophy (Boca Raton, FL: CRC Press, 2009), 100-101.

[5] Allday, 101.

[6] If photons are fired through the experiment one at a time (firing photons at a wall with two holes and a photon detector on the other side of the holes), they will build up an interference patter on the other side, as if they had gone through both holes at once and interfered with themselves.  If the experiment is set up so that detectors monitor which hole the photo goes through, the photon is indeed observed to be going through only one hole, and there is no interference pattern.  If a detector is set up not at the holes but intermediate between the two holes and the back wall detector screen then it may be possible to see which route a particular photon took after it had passed the two holes before it arrived at the screen.  Quantum theory says that if we choose to turn this new detector off and not look at the photons, they will form an interference pattern.  But if we look at the photons to see which hole they went through, even if we look after they have gone through the hole, there will be no interference pattern.  The delayed choice comes into the story because we can make the decision whether or not too look at the photon after the photon has already passed through the hole[s].  The decision made seems to determine how the photon behaved at the time it was passing though the hole a tiny fraction of a second in the past.  It seems as though the photons have some precognition about how the set-up of the experiment will be before it sets out on its journey.  This has also provided credence to the metaphysical concept of backwards causation.  John R. Gribbin, Mary Gribbin, and Jonathan Gribbin (Q Is for Quantum: Particle Physics from A-Z. London: Weidenfeld & Nicolson, 1998), 102-103.

[7] This is most notably accepted by the transactional interpretation of quantum mechanics. Gribbin, 104.

[8] Allday, 102.

[9] Predictability may be equivalent to randomness, not a lack of causality.  Louis Pojman, Philosophy: The Pursuit of Wisdom (Boston, MA: Wadsworth, 2006), 229-230.

[10] Recalling Einstein’s epistemic method, he based all of his philosophy and work on the ontological status of the universe.  He did not seem to indicate an immateriality to the mind.  Einstein’s influence is limited only to the physical aspect for the substance dualist.  Here is where the substance dualist and the scientific theologian must resume the dialogue.


18 Responses to “Einstein on Free Will”

  1. Max, what do you think of theses that use quantum mechanics to argue for freewill but avoid randomness? Like Jeffery Schwartz’s http://amzn.to/nDlxq9 that there could be a “mental force” exuded by our minds (or brains?) that essentially acts like an observer to collapse wave functions on the quantum level thus releasing the ions to make neuron synapses fire? I dunno if it works but it’s really interesting, because it would avoid the randomness objection of using quantum for free will, and also escape determinism.

    • Hey Joshua -

      I agree with you here. I hold basically the same view as Schwartz. It is interesting that you mention him, as I am reading one of his books called ‘You are not Your Brain.’

      I prefer to view the “mental force” as a bilateral interaction that mediates a psychosomatic union. It probably involves an undiscovered interaction with “dark matter.”

  2. Hello Max -

    Just offering an opinion for your consideration.

    More important for one’s view of free will is the model one assumes for the interaction between the mind and the body. If you are of the opinion that the mind is merely an emergent epiphenomenon of the brain, then determinism seems to be your best bet.

    Also, whether or not one is a dualist, the carnal mind (brain) (Romans 8.7) is probably subject to quantum mechanical determinism, which is only a limited form of determinism.

    If on the other hand, you hold the spiritual mind to be an independent entity which has the power to influence or override selection events in the brain, then the outcome of physical brain events may still be determined, but not by solely by pure physical cause and effect. (i.e. agent causation)

    This raises the question of the nature of the interaction between the mind and our neural apparatus. Under what conditions might the natural causal nexus be open to the determining influence of the mind? Since scripture clearly indicates that our spirit is a created entity that is not visible, I propose that the human spirit is a form of “dark matter” that has a physical nature and a hitherto undiscovered force of interaction with ordinary matter. Quantum mechanics may or may not be applicable in the realm of spirit-matter.

  3. I think determinism is still a possibility given QM, but we have to look at the possibility of indeterminism as well. It really depends on which quantum interpretation one subscribes to, and whether one accepts non-local hidden variables to be viable. There are counter-intuitive problems with all of the interpretations, therefore it is best to be agnostic about whether determinism holds or whether there are truly acausal events in the universe.

    Regardless, as you suggested, both scenerios are incompatible with free will. Free will is logically impossible in both a deterministic universe as well as an indeterministic universe (I hold the hard incompatibilist stance rather than the hard determinist stance). I am currently writing a book explaining in detail why both of the only two possibilities (events being causal or events not being causal) do not allow for the free will that grants responsibility, and why it is so important for people to understand this fact.

    Take care,
    ‘Trick Slattery
    http://www.TrickSlattery.com
    http://www.breakingthefreewillillusion.com

  4. ‘Trick –

    Actually, neither scenario is incompatible with free will, properly understood.

    Quantum mechanics actually makes room for selection events, because quantum determinacy is not absolute. It is properly viewed as limited determinacy because definite states are limited, not uniquely specified by the evolution equation. The wave equations evolve deterministically, yes. But measurements involve actualization of one state out of multiple possibilities. These measurements should be viewed as ‘selection events’ and they provide room for one or more selection factors. Thus, limited determinism of selection events in QM systems leaves room for intervening agency to act deterministically, according to a freely acting will.

    • Hiya Kevin,

      When you say “properly understood”, I wonder what you mean by that. The free will definition that concerns me is the one that implies responsibility.

      Also, your statement “But measurements involve actualization of one state out of multiple possibilities” is of concern. They can only be possibilities if indeterminism holds (ie. acausal events happen). If the universe is entirely causal (deterministic), there can only be one possibility that can ever come to be. If there are acausal events (the universe is indeterministic), no acausal event can be a willed event as wiling requires a cause (the willer). There really is no room for freely acting will.

      Take care,
      ‘Trick

      • Hello ‘Trick –

        Your analysis is overly simplistic, in that it doesn’t take into account the nature of QM “determinism” and how that creates space for interaction of an external agent with the neural apparatus. By “properly understood” I refer to the nature of agent causation and QM determinism.

        (1) To be “free” I take to mean that for the agent, willing acceptable outcomes is at least a live option. In other words, willing is not mechanical. It is deliberative. Also, an outcome of any specific moral choice can be subjected to one’s higher order preferences, which are themselves at least indirectly selected by the agent. I agree that the critical issue is responsibility. Without freedom to determine the outcome of a morally significant choice, there can be no responsibility, that is, unless one’s inability is itself the result of culpable negligence.

        (2) You said, “They can only be possibilities if indeterminism holds.” This is not true. It misunderstands the kind of determinism in QM. As I alluded to before, the Schroedinger equation yields a specific set of solutions that are deterministic in the sense that the solution set is determined to be limited to specific eigenfunctions. QM limits the possibilities, but does not always or even usually result in a uniquely determined outcome. A simple example would be degenerate orbitals of an electron in an atom.

        Multiple possibilities can obtain in the absence of a chooser. In other words, in a selection event, in the absence of an interacting agent, the state of the system is not completely specified. It is the action of the agent that results in the determination of the outcome. This is agent causation.

        If moral decisions involve such quantum selection events within the neural matrix, there is room for limited freedom corresponding to a limited set of QM states of the system. Choices are usually like that. The question then is whether or not the agent is governed by the laws of QM.

        My view is that the brain and the mind are interacting but independent systems governed by different rules. And I believe this view is supported by clinical research on near-death experiences.

      • Hi Kevin,

        I apologize if this is too long.

        We haven’t really delved into my “analysis” ;) The book I am writing has chapters devoted to the misunderstanding of QM in regards to the free will debate.

        “To be “free” I take to mean that for the agent, willing acceptable outcomes is at least a live option.”

        I would only consider this “willing” not “freely willing”. I think causality is entirely compatible with a person willing, in fact, it is a necessity to willing. Where the mistake lies is thinking that “higher order preferences” do not come about via complex causal chains that could have happened in no other way (without acausal events). Such “higher order” events stem to events that are entirely out of the persons control (lower order, or not even within the confounds of the mind making the decision).

        “It misunderstands the kind of determinism in QM.”

        I am not understanding what you mean by “kinds of determinism”. Determinism simply means that every event has a cause. There are “kinds of causality”, but not “kinds of determinism”. For example, if one accepts a Bohmian Mechanical interpretation of QM, one must accept non-local causality. Regardless, the same cause cannot (logically) have multiple possible effects. Cause X cannot be the cause for both effect Y over Z as well as effect Z over Y. To be led to other than the one dictated by the cause, an acausal event is required.

        “in the absence of an interacting agent, the state of the system is not completely specified”

        Given that we accept this interpretation (that until measured, the state is between “possibilities”), once an interaction happens is when a cause happens for it. A cause that can only force one very specific “possibility” from the reservoir based on the variables of the cause. We can never have an interaction in which such interaction does not cause what its variables lead to. If determinism holds, there really is no “multiple possibilities”. There is only the one possibility dictated on the interaction, which such interaction is dictated by the variable that precedes it. If we suggest otherwise, we must suggest that it is an acausal event that takes place to determine which “possibility” comes to the forefront, not a causal event.

        There are so many different competing Quantum interpretations. Each of them run into this problem. For those that are deterministic such as many worlds or hidden variable interpretations, there is no way around the logic for this. For those that are indeterministic such as Copenhagen, we must contend with the nature of acausal events (events that do not have a cause – including the willer). In other words, even if the agent is governed by QM, such a state would be entirely incompatible with free will.

        Thanks,
        ‘Trick

  5. Yeah Trick, you could be barking up the wrong tree. I think Nancy Murphy and Warren Brown are right that the entire free will debate has been misframed: Determinism isn’t the problem, reductionism is the problem. I think they’ve done a good job showing you can have libertarian free will along with determinism. http://amzn.to/oFKno3

    • Hi Joshua,

      A reductionistic framework is not a necessity for the free will debate. Downward causation is equally incompatible and so is higher order causation. Multiple heirarchies of causation to not help grant free will. I go over this in the book I am writing as an “anti-reductionistic framework” in no way helps support the notion of free will. It is a large mistake to think the hard determinist or hard incompatibilist are such due to reductionism.

      Thanks,
      ‘Trick

  6. Trackbacks

Leave a Reply