Recently, co-blogger David Henderson offered some ponderings about the limits of self-ownership. He argued that the government shouldn’t put limits on the self-ownership of adults of sound mind, but such limits could appropriately be placed on children and adults of unsound mind. That raised the issue of how we set about determining when a child stops being a child and counts as an adult, or at what level an adult is of unsound mind.
Of course, through policy we can set particular hard and well-defined limits. If you’re over 18 years of age, you’re an adult. If you’re younger, you’re a child. If your IQ is above, say, 85, you’re of sound mind, but if it’s below that, then you’re not. This approach is clean, neat, simple, and as David points out, comes with two big difficulties.
One difficulty is that drawing a hard line at a specific threshold will always be arbitrary to a certain degree. In day-to-day life, the mental capacity of someone with an IQ of 85 is basically the same as someone with an IQ of 84. Yet the former will be classified as of sound mind, while the latter is considered of unsound mind, despite there being no real practical difference between them. Someone who is 17 years and 364 days old isn’t going to change in any meaningful way when the clock strike midnight and they turn 18 – but should they happen to have been charged with a heinous crime, whether they get charged as a minor or an adult can very well turn on if the act occurred before or after midnight.
The second problem is that there will always be cases on either side of the line that are wrongly classified. There are some people with an IQ of 80 who, despite not being intellectual powerhouses, can still conduct themselves well and make sound decisions. And there are people with an IQ of 180 who go mentally off the rails and make nutty decisions. I’ve known people who at 16 years old were more mature and insightful than some 26 year-olds.
How do we solve these two difficulties when it comes to establishing limits? If you were hoping I’d be solving that problem, I’m going to have to disappoint you. But I was reminded of an insight from Daniel Dennett’s book Intuition Pumps and Other Tools for Thinking where he described how we can tie ourselves up with this question.
He begins by offering a simple argument that mammals don’t exist, starting with two premises that initially seem plausible:
- Every mammal has a mammal for a mother.
- If there have been any mammals at all, there have only been a finite number of mammals.
- But if there has been even one mammal, then by (1) there have been an infinity of mammals, which contradicts (2), so there can’t have been any mammals. It’s a contradiction in terms.
Of course, Dennett isn’t seriously suggesting there are no mammals. He instead invokes a Moorean shift, arguing that since the conclusion is obviously absurd we can justly conclude that something is wrong with the premises or the inference, thus “we take this argument seriously only as a challenge to discover what fallacy is lurking within it.”
One way we are tempted to solve the issue is by disputing the first premise. Along the evolutionary path, there were precursors to modern mammals called therapsids, animals that weren’t quite reptiles but not really mammals yet either. So, Dennett suggests, perhaps along this line we could, in principle, identify the birth of what he calls the Prime Mammal – the first animal born that was Objectively And For Real A Mammal. Thus, when this Prime Mammal was born, we have a mammal that was born from a therapsid, falsifying premise one and preserving the existence of mammals. Hooray!
Except, hold on, it’s not that simple. Let’s just imagine we had godlike technology allowing us to look back through time to try to spot the birth of this Prime Mammal. When this Prime Mammal was born, how would we know it? What would be the characteristics that make it essentially and fundamentally unlike its therapsid parents? And here we run into the problem of arbitrariness. We could just stipulate that, perhaps, there are ten features that define a mammal, and the first animal born with all ten features is the Prime Mammal. But that’s clearly arbitrary. Why ten features? Why not eight, or seventeen? And why that particular set of ten and not some other set of ten? And along the evolutionary chain, we would find all kinds of instances where a specific animal with ten of those features (and therefore a mammal) mated with another that had only nine (and therefore a therapsid), producing new generations of mammals born of therapsids alongside therapsids born of mammals.
Dennett suggests we handle this problem in a way that would probably make many philosophers very upset – we deal with it by not dealing with it:
What should we do? We should quell our desire to draw lines. We don’t need to draw lines. We can live with the quite unshocking and unmysterious fact that all of these gradual changes accumulated over many millions of years and eventually produced undeniable mammals. Similarly, the differences between lakes, ponds, and wetland or marshes do not need to be calibrated, even by limnologists (those who study inland waters).
This runs counter to how philosophers like to think about things:
Philosophers, however, tend to be tidy, fussy users of words. Ever since Socrates persisted in demanding to be told precisely what marked the defining features of virtue, knowledge, courage, and the like, philosophers have been tempted by the idea of stopping a threatened infinite regress like this one by identifying something that is – must be – the regress-stopper: the Prime Mammal, in this case…So, as a general rule, consider ignoring the philosophers’ demand for an essence, a defining feature, a “truth-maker.” It typically – not always – starts a wild goose chase that may be diverting but is only moderately illuminating at best.
The upshot – the act of where and how to set limits is itself an activity that has limits. We shouldn’t worry too much about our inability to set exact limits in a way that correctly handle every case, because that’s simply not a possible task. (As I once heard a mathematician quip, it’s very difficult to prove a theorem that’s false!) Now for some things, we simply can’t take Dennett’s advice to just avoid drawing any lines at all. Attempting to solve everything on a case-by-base basis has prohibitively high transaction costs. But when we draw these lines, we should do so with the full awareness that any line we draw will necessarily be imperfect and will get things wrong some of the time. (This, perhaps, is another reason to ensure rules allow for the possibility of exceptions.)
However, one mustn’t make a different sort of mistake from this – the mistake of going from the observation that there is no objectively correct, nonarbitrary spot to draw a line for these distinctions, to concluding that the distinctions themselves are meaningless, anything goes, and it doesn’t matter at all where the line ends up. Matt Zwolinski talked about how people can make this mistake when they move from pointing out that certain features of property rights require enforcement through social conventions that are inevitably arbitrary regarding specific details, to embracing the non-sequitur that the very idea of property rights is arbitrary and entirely determined by social conventions.
As Zwolinski put it,
One way of sidestepping the philosophical puzzles involved in these questions is to simply stipulate an answer through the establishment of a convention. The Homestead Act of 1862, for instance, stated that families could claim up to 160 acres of land once they’d lived on it for five years. Why 160 and not 180? Why five years and not three? Obviously it’s not because those numbers are uniquely mandated by the correct theory of natural rights.
Does that make the theory of natural rights worthless? Of course not. A theory of natural rights establishes general principles, and those principles demarcate a range of morally acceptable solutions to the problem of appropriation. Within that range, societies are free to choose. But the open-endedness is not unlimited. It doesn’t really matter if you specify that families can claim 160 acres or 180. It matters a lot that you specify that they can’t take land that somebody else is already living on.
So what does all of this mean? It’s true that there is no one, unique set of rules and institutions that counts as a “free market.” Neither natural law nor economic theory can tell us exactly what a libertarian utopia should look like. But that doesn’t mean that anything goes. It might be impossible to specify in a non-arbitrary way exactly where blueish-green shades into greenish-blue, or when a child becomes an adult, or when a free-market ceases to become free. But only someone who has allowed philosophical puzzles to blind them to the world in front of them would conclude from this that there’s no difference between green and blue, or a child or an adult, or capitalism and socialism.
The same, too, can be said of someone who insists that there is no difference between a mammal and a reptile. We can’t draw an exact line justifiable with mathematical precision. The best we can do is find a point in the proverbial grey area, draw a line somewhere in there, and say “well, that’s good enough.” And in much of life, “good enough” is going to be the best we can do.
