unwinnable conditions in the game of defining free will

The sad truth is that games can slip into unwinnable conditions: game situations that cannot lead to an eventual victory. Many one-player card games routinely become unwinnable through an unlucky order of cards in a stack...or short-sighted mistakes in how the player moved the cards around. Some two-player games with multiple pieces, such as chess and checkers, become unwinnable by either player once they're left with next to nothing on the board. Adventure-style computer games, which depend on collecting and cleverly using items to escape danger, become unwinnable through losing (or overlooking) items that will be surprisingly crucial to survival.

Philosophical debates have a tendency to become unwinnable too. This happens when precious yet poorly defined ideas are somehow expected to meet clashing requirements. The result is that the corresponding debates cannot be won, if "winning" means hammering out a self-consistent version of the idea that's satisfactory to everyone in the debates. But the proper response isn't giving up. It's returning the debate to a winnable condition by dropping or reshaping some requirements. 

The idea of free will is a prime example, since discussions about it frequently end in stalemates. And it's particularly appropriate here because game-playing itself is an analogy for it. Regardless of how relatively trivial a game might be, the players carry out the widely recognized aspects of free will. They analyze the circumstances, then select from a set of actions, to attain goals. (Of course, topics have been usefully compared to games for a long time. Wittgenstein described language games. Game theory is a branch of mathematics that's constantly referenced in a range of contexts.)

Some common unwinnable conditions of the effort to define free will are worth examining. To echo the statement from earlier, the aim of doing so isn't to abandon free will but to pinpoint one or more troublesome requirements. The most fundamental of them should be attacked first: that free will needs to be capable of violating the physical laws that are followed by things that don't have free will. In a word it needs to be unabashedly miraculous. The equivalent in gameplay is spontaneously disobeying the game's rules—which is not the same as when all players agree upfront to slight modifications.

This places the debate into an unwinnable condition because the best evidence available simply hasn't uncovered this kind of violation. At the scale of the person-centered decisions customarily placed in the category of free will, the rules are firm and uncontroversial. Atoms aren't created or completely destroyed, although some decay. Energy only changes form, and it tends to become more diluted when it does. Velocity stays the same unless something acts to change it—including a velocity of zero. Objects cannot be accelerated to the speed of light. Differing electric charges are attracted. If free will doesn't follow the known rules or patterns, then it cannot be reasonably integrated to the rest of reality. And presumably it could contradict any precise definition assigned to it.

A believable idea of free will should stay within the same broad boundaries that allow for the behavior of galaxies, continents, algae, clouds, platypi, etc. Furthermore, there's an upside: reliable consistency. This is crucial for effective short-term decisions and also long-term plans. Acquired items don't vanish. Notebook pages filled with ink don't switch back to blankness. Projectiles descend at an expected speed. Food temperature doesn't suddenly diverge from the environment it's in. Arguably, without impartial rules to govern the consequences of actions, free will wouldn't be worth a lot in practice. The rules are tools as well as limits. People are participants in rule-governed existence, not spectators.

Admittedly, committing crime against physical laws is extreme. Most probably don't lump together free will and sorcery. By comparison, a seemingly moderate compromise is imagining the power to just disconnect from potential influences. Through the application of this power, decisions could be rendered perfectly independent, separated from external things, internal pressures, or the past. Decisions in this state could not possibly be manipulated by anything.

Nevertheless, this requirement's drastic solution has two problems. The first problem is that it leaves decision-making with a hollow core. It's certainly good that nothing can indirectly yank around decisions like pulling the strings of a puppet. But if the decision isn't determined by anything, then it's uninformed. It's cut off from the context that provides meaning to why action A was selected instead of B. If it's made for no deeper reason, then it's due more to chance than thoughtfulness. High unpredictability isn't synonymous with willful freedom. The problem isn't eliminated by merely using a fancy source of randomness, such as the precise probabilities of quantum mechanics. A heavily randomized decision is still the opposite of taking responsibility. It's more like the game act of rolling dice than the act of choosing to skip an optional dice roll or not. (Or it's like flipping a coin to choose to roll the dice?...)

The second problem with free will hinging on the power to disconnect is the sheer implausibility. Modern knowledge shows, in addition to the previously mentioned physical rules at the human scale, a grand web of cause and effect between the multitude of particles and energy fields of the universe. Therefore it's frankly bizarre to picture a tiny strand wandering off whenever the urge strikes. How could it possibly do that? Why should it have such authority over other strands joined to it? Breaking cause and effect at will qualifies as a superpower. Demanding it isn't a winning approach for arguing that free will is realistic.

An alternative is therefore necessary. Adaptation fits the role. It's when actions vary based on influences. The definition is vague because the actions fall into a host of categories. Adaptation might consist of taking the influence's impact and neutralizing/dampening, or amplifying, or aggressively countering, or preventing, or redirecting. In general terms it's perception followed by corresponding action. 

Importantly, adaptation in the vague sense isn't a unique characteristic of human consciousness. It appears in many things of differing complexity. Single-celled organisms react depending on events in their minute but brutal ecosystems. Computer systems jump between Wi-Fi routers. Hedgehogs roll up into balls. Metaphorically, a spring could be said to adapt, because its "actions" vary based on the stress of squeezing. Obviously, its adaptation is far different from, say, people adapting to their circumstances by revising their life insurance. The part that matters is that feedback, the primitive building block of adaptation, isn't unusual or ghostly at all.

Moreover, advanced multi-layered adapting can come quite close to gaining effective control over decisions. If multiple influences are observed, understood, and consciously managed, then the influences are mastered. No single influence dictates the decision. The decision-maker isn't metaphysically separated from influences, but they don't need to be. All the paths from influences to the decision can be, er, adapted: continuing to exist but now much more complex. Within a person, these adapted paths are literal. The signals from the influences take more complicated routes through the brain's network of cells. 

For one person, the path from stimulus to rage (or panic or gloom, etc.) could be extremely simple and predictable. For another who has adapted by noticing the first signs of oncoming rage, and who has spent significant time weighing the pluses and minuses of how they act, the path traveled by the stimulus could be twisty and stop in any of several outcomes. For the latter, the "natural" or quicker path hasn't been vaporized by psychic force. It's been circumvented through the development and activation of competing paths. And since everything is material, the process of overriding costs actual energy. It's like two crowds attempting to outshout each other. With this in mind, free will can be more of an ideal to strenuously pursue than an inborn capability. 

One last feature deserves mention. As a temporary adaptation of behavior or thought is repeated over time, it's reinforced. This is an aspect of how brains evolved to learn. And with enough reinforcement, the corresponding brain path functions as the relatively "natural" or quicker one. So adaptation may become a method of self-change and even a type of self-determination. It has the caveat that it's gradual and difficult. Uttering the magic spell "I am different" doesn't instantly engrave adaptations into the self, any more than uttering a magic spell doesn't instantly etch a design into stone. It calls for visionary thinking and persistence, a combination which few things are capable of. To become an expert at a game, nothing helps as much as the "work" of playing countless times, making good and bad moves, in order to replace beginner instincts with seasoned ones.    

Excessive focus on self-governance has a subtle risk. It might evoke a requirement that makes a definition of free will unwinnable: treating denial as too pivotal. To suggest that rejecting an impulse reflects free will, but fulfilling an impulse doesn't, runs into trouble. First, if denial isn't serving the purpose of a desired result, then it's not inherently meaningful. That purpose might be nothing beyond "proving that I'm the sort who doesn't do that" or "getting some enjoyment out of punishing myself", but it's present. Denial isn't a self-sufficient motivation. Second, the statement can be flipped. A decision-maker who's absolutely ruled by denial isn't convincingly freer than a decision-maker who's absolutely ruled by fulfillment. 

Third, some cases don't involve a strong element of denial in the outcome. Yet the decision-making process that led to fulfillment could be as rigorous as a process that led to denial. In either decision, the influences could've been thoroughly mastered (...or not). Picking an influence to obey can be a free choice. Free will isn't viewing decisions as stark intellectual puzzles, after pushy motivations have been completely drained out. Relative preferences at the very least should be factored in. If someone were playing a game, nobody would maintain that a craving for victory is incompatible with a reasonable style of play.

Clearly, a kind of free will can function without dragging along unsupported assumptions. That might not stop especially stubborn debaters from insisting on unwinnable conditions anyway, by raising a final requirement that to them seems equally clear. As they would say: for people to be truly free, they must make use of something apart from particles to make decisions. After all, a particle doesn't have the freedom to deviate from the predictions of a skilled onlooker. Predictable particles adding up to a person with unpredictable free will doesn't make common sense.

The retort won't surprise anyone who's heard this before. Despite their particles, people are in fact complicated. Particles connected in certain ways can rapidly "add up" to more unpredictability than one in isolation. Once a particle is connected to its surroundings, then predicting the particle depends on predicting the effect the surroundings will have on it...and that means predicting the movements of these surroundings. Then those surroundings in turn cannot be predicted without predicting the outer surroundings that affect the closer surroundings. The mandatory edges of the predictions keep getting pushed farther and farther. 

For instance, the full motion of a particle in one of the bones in one of the fingers of a person typing an email isn't predictable unless the next letter to be typed is predicted. The next letter depends on the word being typed. If the email is about an amusing memory gleaned from the previous day, then the word being typed depends on the person's mental reconstruction of the past. Or they may jump up from the keyboard after the doorbell rings, and then the motion of the one particle can't have been predicted without also predicting the doorbell ringer.  

The essential problem is there are too many particles interacting and too many independent motions to track. Transforming these inputs into accurate particle predictions would consume an impractical amount of space and time. And the dense simulation might start failing to match reality less than five minutes later, when the person receives a text asking for help...sent from someone on the other side of town. Sure, human free will without something apart from particles could be calculated, but it's pure fantasy. And it's irrelevant next to the immense realm of possible actions for people made of particles, as they sift through all of their intricate concerns and justifications and construct endlessly creative variations. Given that deceptively limited games have turned out to have fascinating depths of strategy, there's no need to worry about feeling restricted by the few real walls around human decisions.

(Addendum: A short while ago I read How Physics Makes Us Free by J. T. Ismael. Parts of this blog entry are almost certainly surface-level restatements of ideas I picked up in it. But the book covers interesting topics such as the formation of coherent self-voices, entropy in human experience, and the network model of causes and effects. It also teaches words like diachronic and modal and immanent, with a glossary in the back.)