Model Dependent Realism, as articulated by Hawking and Mlodinow (2010) and Carroll (2016), is the idea that a specific and explanatory model represents a ‘reality’ and is ‘true’ if it can both account for extant data and is predictive. I have used this philosophy before to contest that determinism is the explanatory model that best accounts for data about the brain when making decisions (2017). I now want to argue something frustratingly contrary to that thesis: freewill is the better predictive model.
I should start by getting some things out of the way: freewill is not a particularly specific model, it lacks any explanation or mechanism. That does weaken its standing in a model dependent realist approach. In fact, I think the thesis I have written here fails to generate a good defence of freewill, the merit I find in writing it is the hope is sparks a worthwhile conversation or continues to demonstrate that freewill really does fail to have an intellectual defence. Run with the discussion for now.
It is worth pointing out that the assumption of freewill does a better job of predicting complicated decisions. Simple decisions are predicted by determinism (Soon et al., 2008) and it may be that it only requires an advance in technology to show that complex decisions are also predictable under determinism. But, as it currently stands, that is not the case.
It is easy, and even understandable, at this point to ignore freewill altogether because Soon et al has demonstrated that simple decisions are predictable and there is no principle difference between a simple and a complex decision. However, operating under a wise cautionary philosophy discussed by Deutsch (2011) ― a sort of knowledgeable realism ― we must operate within the confines of what we know and keep our extrapolations and conjectures as provisional. We may be able to conjecture that the simple decision is perfectly scalable to the complex decision, and therefore Soon et al’s work is applicable. But for the sake of intellectual caution, let’s not be sure of that jump, yet.
The mind is chaotic. I mean this is the technical sense mathematicians use it: sensitive to initial conditions (Boeing, 2015). It does not mean the mind is, in principle, unpredictable; but neither does it exclude that. It means that the slightest variation in how the mind starts a particular event has much larger and more profound implications on the outcome. The consequence of this is that, at present, the mind is practically unpredictable. And, for the same reasoning, as well as the fact the two are very closely linked, so is the brain. Given the unpredictability of the mind and the brain, determinism begins to fail one of the criteria of model dependent realism: making predictions of complicated decisions.
However, that is not the same as saying that people’s minds are unpredictable. We can, with some reliability, make predictions about how people we know well will make decisions. We know their values and their virtues; their influences and what they plan to avoid; we know them in the ways that are important. Psychologists can make the decisions very reliably, using models that assume decisions making and freewill.
There is an important question raised here: if a perfect psychologist is capable of predicting your decisions, does that not assume some sort of fatalism or determinism? It may not be the neuroscientific determinism I argued a defence of previously, but surely having predictable decision-making still implies determinism over freedom. There is probably a lot of work to be done here, still, on making ‘freewill’ a better defined and more specific model. That’s another hole in this thesis attempting to support freewill.
However, in looking at how we predict our friends’ and family’s decisions, and how a psychologist make predict the decisions of an addict, we see that our kind of reliable method assumes freewill. It seems self-contradictory (and I’d argue that it is contradictory) that assuming freewill makes people’s decisions predictable.
So, there’s the best thesis I can honestly muster in defence of freewill: when one focuses only on complex decisions, assuming there is no scalability between simple and complex decisions, one makes a better job of predicting people’s decisions knowing the personal profile of an individual and assuming freewill. However, this very conclusion is paradoxical because it assumes that freewill is better explained by having predictable decisions. Other pitfalls in the idea of freewill are around how well defined it is and whether there is any explanation of how it might work.
In summary, I cannot see how honestly evaluating the idea’s truth leads one to accept it. It may be that one accepts it for values on how to navigate politics and ethics (and one may even defend that as a ‘pragmatic truth’), but in terms of an actually ontological reality, I cannot see how it can be.
This is my first post written and posted using the WordPress Add-on for Google Docs. My thoughts on it are that it is convenient, but there is a lot more it could do. For example, the menu on the right hand side of the Docs screen could include a window for adding tags and scheduling.
I cite online versions of books, which is why they are referenced without cities of publication.)