The Lure of Compatibilism
The Lure of Compatibilism
Why Most Philosophers Think That We Have Free Will Even if Everything is Already Determined
Most philosophers are compatibilists: they don’t believe that determinism and free will are mutually exclusive. So, even if we reside in a world wherein the future and past are fixed, and even if somebody with enough data and compute could predict your behavior, and even if neuroscientists monitoring your brain could anticipate your decision-making, compatibilists still think that free will is possible. In short, even if your actions are completely entailed by history and the laws of nature, compatibilists argue this wouldn’t preclude free will. It’s an outlandish position. But it’s also the most common view among philosophers—not just the abstruse plaything of a few heterodox weirdos. For me, reconciling the discrepancy between compatibilism’s manifest implausibility and its longstanding popularity has always been the most frustrating aspect of studying free will. It’s a persistent challenge for me to assign charitable interpretations to compatibilist ideas rather than dismiss their entire enterprise as casuistry and denialism.
Abandoning Classical Compatibilism and Redefining Free Will
The traditional argument for compatibilism is that determinism and free will can peaceably coexist because even if someone could not have done otherwise in a strictly identical world (given determinism), they usually could have done otherwise if they had wanted to. This reasoning is simple and, prima facie, intuitively compelling: it captures a distinction that’s tempting to characterize as free will. As I’ve written before, however, this style of compatibilism has pretty much been debunked, since someone wanting to do otherwise is just as susceptible to determinism, and so redirecting the origin of freedom to whatever someone wanted to do ultimately doesn’t accomplish very much, philosophically. Or, as I’ve put it before, “[W]ho gives a shit if you wanted to do otherwise when what you wanted to do was also predetermined?”
Harry Frankfurt’s “Alternate Possibilities and Moral Responsibility” ushered in a major strategic evolution for compatibilists, who started employing Frankfurtian Counterexamples to question whether free will and moral responsibility required alternative possibilities at all (instead of trying to preserve some ersatz notion of alternatives within a deterministic world). Here’s an example:
Frankfurt Counterexample: Suppose Alice has free will and an ability to do otherwise, and she’s planning to kill Bob. The evil neuroscientist Charlie has a vendetta against Bob and is counting on Alice to assassinate him. To ensure Alice goes through with it, Charlie installed a device in Alice’s brain that will override any attempt by her to do otherwise. Alice assassinates Bob and the device never intervenes.1
And here’s the explanation of a similar example that I gave before:
In examples like this, Alice couldn’t have done otherwise (because of Charlie’s device), but this doesn’t seem to let Alice off the hook: Charlie’s device was irrelevant to what transpired. The fact that she couldn’t do otherwise wasn’t the reason why she killed Bob. . . . Alice would have killed Bob even without Charlie’s device and be blameworthy, and the device didn’t interfere at all, so why would that device’s presence suddenly relieve Alice of being blameworthy?
But this maneuver also requires recharacterizing free will as some notion of autonomy besides the power to do otherwise. So, it potentially involve some conceptual revision.2 Put differently:
Sure, compatibilists believe in free will, but the compatibilists’ idea of free will is a kind that’s accessible even when your every move is predetermined, and modern compatibilists often (mainly?) believe that free will and determinism can co-exist only because they think that free will is something that doesn’t include the ability to do otherwise. Crucially, this is all pretty opaque to outsiders, and maybe non-specialists would view this as a surprisingly modest conception of free will. Thus, whenever you hear something like that most philosophers believe in free will (because most of them are compatibilists), take it with a grain of salt.
So, while it’s arguably warranted, there’s definitely some goalpost moving here. Also, while I’m a non-expert about the history of this transition in strategies, I've always thought it was pretty telling how compatibilists gladly switched gears and abandoned their traditional line of reasoning—like, maybe their whole project is just rationalizing their instinctively preferred viewpoint.
Actual Sequence Compatibilism
While Frankfurt Counterexamples are undoubtedly clever, I believe that they’re fatally defective.3 For example, Alice could have started doing otherwise before triggering Charlie’s neural hijacking—she could have done otherwise just a little bit. Ultimately, though, there’s controversy about whether these Frankfurt Counterexamples need to be metaphysically flawless to be rhetorically legitimate. Do they still showcase something important even if they’re doomed to imperfection? Frankfurt apologists contend that such feeble alternatives (like Alice starting to do otherwise) are too insubstantial to factor into assessments of freedom and desert. Maybe philosophers became distracted by the flashy thought experiments and lost track of the underlying rationale.
Frankfurt’s fundamental insight was that the reasons someone can’t avoid X aren’t necessarily the same reasons motivating someone to do X. So, one way to digest these flawed counterexamples is to focus on the intuition that moral responsibility turns on what actually happened rather than what could have happened; it’s an invitation to shift your thinking about agency from an alternative possibilities (”AP”) framework to an actual sequence (”AS”) framework. Carolina Sartorio nicely encapsulates the competing instincts like this:
The AP model is plausible, among other things, because it is natural to think of freedom in terms of alternatives, and the AS model is plausible because it is natural to think that only factors that actually explain why an agent acted can be relevant to the agent’s freedom.4
Thus, I might charitably portray the idea of AS-compatibilism as being that there’s some kind of significant volition/control of one’s actions based in the causal history of those actions rather than solely on the menu of actions available for selection.
The Scientific Image and The Manifest Image
Fellow substacker
subscribes to compatibilism based on some other ideas and submitted a valiant effort at animating those intuitions in his recent post. He doesn’t use this terminology, but his primary argument resembles Wilfred Sellars’ famous distinction between the Scientific Image and the Manifest Image. The basic idea is that the Manifest Image (the world as we normally observe it) and the Scientific Image (the world as scrutinized by the refined analyses of modern science) offer dissimilar-but-equally-valid perspectives of reality: the same phenomenon can be understood usefully at varying levels of resolution, and it’s erroneous to presume that incongruities between those perspectives result from genuine disagreements rather than arising naturally from differences between the projects they’re undertaking.Blanchard uses the example of a calculator: you can explain its outputs either in terms of its electronic inner workings or in terms of arithmetic, depending on the situation—both explanatory strategies are legitimate. Daniel Dennett also championed the relevance of Sellars’ bifurcation of the Scientific and Manifest Images to the concept of free will. For these Sellars-acolytes, thinking that free will is disproved by granular scientific experimentation is an error of myopia, like when people confusedly aver that paper money isn’t real because it doesn’t have intrinsic value (as opposed to something like gold). Like your bank account, free will is real, but it can’t be isolated and scrutinized like a bacterium (so they claim), and we don’t consider the macroscopic characteristics and functions of everyday objects to be figments of scientific ignorance merely because they look radically different under the microscope. Thus, someone like Robert Sapolsky, by trying to disprove free will via neuroscience, has conflated the Scientific Image and the Manifest Image.
While Sellars’ distinction is valuable, it doesn’t fully defuse the scientific threat to free will (let alone the myriad philosophical quandaries), and I disagree with the claim that free will skepticism is just a confusion resulting from competing tiers of inquiry. The commonplace Manifest Image of free will comprises attributes that are fundamentally inconsistent with the Scientific Image’s more detailed examination—they’re mutually exclusive, and ultimately the Scientific Image must take precedence.
Freedom, More or Less
Blanchard nearly bumps against another gambit for jogging compatibilist sympathies when he inquires how we should distinguish between actions like winking and blinking, because while a free will skeptic can sensibly distinguish between the two behaviors, they cannot do so based on any philosophically meaningful sense of freedom/agency that isn’t, at bottom, illusory. To put it differently, those who doubt free will are committed, a fortiori, to rejecting any comparison between actions claiming that one is done more freely than the other: even if winking is more voluntary than blinking, it must be equally unfree.
The skepticism that commits you to the idea that free will is a complete nonstarter likewise commits you to a lemma that nothing can be done more freely than anything else, since nothing can be done freely at all. In some sense then, for the free will skeptic, that everyday impression of voluntariness is misleading, and ultimately there’s not much difference between stuff like winking and blinking outside of utilitarian concerns about behavioral incentives: you couldn’t really blame someone for something volitional like winking anymore than you could blame them for something involuntary like blinking.
Even if the vague psychological impression of autonomy that we enjoy doesn’t present a convincing response to the gallimaufry of counterarguments, science, and unyielding logic of free will skepticism, this question of whether some actions can be more or less free than others throws the competing intuitions at play into high relief. It’s easy for me to imagine how this appears to present a philosophical misstep—that free will skepticism must be failing to capture some crucial feature of human action. I’m willing to bite the bullet here, but it’s understandable why some could more readily view this outlook as the signature of faulty reasoning than the culmination of shocking, airtight logical deduction.
Conclusion
The arguments against free will are both legion and persuasive: upon delicate consideration, the universe looks inhospitable to anything like it, but compatibilists disagree, and I endorse Blanchard’s view that the intuitions driving that disagreement merit better delineation. To me, compatibilism has often looked somewhat circular in shape, justifying our intuitions about free will’s existence by appealing to our intuitions about free will, basically. I guess the arguments were supposed to be obvious or self-evident, but some more concrete justifications do exist beyond that circularity. The pas de deux of redefining free will as the agency needed for moral responsibility and then conditioning blameworthiness on the actual sequence of events is a nuanced-but-illuminating procedure, and the potentiality of having to ditch the concept of varying degrees of freedom is worth mulling over. Thus, even though grokking the motivations for compatibilism is an especially challenging undertaking for some of us, excavating carefully enough reveals there are some legit intuitions and arguments underneath.
Whether such a revision is so severe that it overturns the pre-existing concept is an independently confusing topic altogether.
Many disagree. The literature on Frankfurt Counterexamples is mind-blowingly complex and voluminous, and to paraphrase Fermat, this margin is too small to contain my proofs.
Sartorio, Carolina. “A Partial Defense of the Actual-Sequence Model of Freedom,” Journal of Ethics, 20 (2016): 112.
This Blanchard guy sounds cool
> As I’ve written before, however, this style of compatibilism has pretty much been debunked, since someone wanting to do otherwise is just as susceptible to determinism, and so redirecting the origin of freedom to whatever someone wanted to do ultimately doesn’t accomplish very much, philosophically. Or, as I’ve put it before, “[W]ho gives a shit if you wanted to do otherwise when what you wanted to do was also predetermined?”
I don't think this debunks classical compatibilism at all. In fact, it's just a circular argument. Who cares if what you wanted to do is predetermined? The entire conceit of classical compatibilism is that this doesn't matter.
Classical compatibilism doesn't hold that the origin of freedom is in being free to want certain things - it holds that the origin of freedom is in being able to do what you want, regardless of how what you want was determined.