There is a great deal to say about Molyneux’s The Art of the Argument, none of it good. The book begins with an introduction to logic that misnames (and misapplies) core concepts like validity, truth, and soundness (Loc 213), confusing how the “truth” relation is supposed to work (Loc 1572-1579), among other basic technical errors. I could dedicate an entire review to showcasing the bizarre and confused claims of the first quarter of the book; when I mentioned reading it on twitter, however, another graduate student reached out to me that he was going to be doing that review. Instead, I thought I’d do something a bit different (and, I hope, a bit more interesting).
In his opening moves, Molyneux differentiates between “Truth arguments” (Loc 68) and “Value arguments” (Loc 70). This roughly maps on to a philosophical distinction between descriptive and normative claims. “The window is open,” is a descriptive claim. “Close the window” is a normative claim. Some philosophers think that some descriptive claims may be normative, and vice versa. (For example, when my girlfriend gives me a look and says, “the window is open,” she is actually directing me to do something about it.)
In keeping with this bit of Molyneux’s analysis, I want to argue that the core argument of The Art of the Argument deserves both a descriptive and normative note (respectively): the argument fails and one who is committed to the argument ought to reject Molyneux’s views about libertarianism. Put another way, the argument is both wrong (a description) and those who accept it ought to reject Molyneux’s libertarianism. (Of course, I think everyone ought to reject libertarianism, but that requires responding to more serious political philosophers than Molyneux.)
Molyneux says we have an obligation to ostracize those who fail to meet certain philosophical standards (Loc 615-630). If I agreed on that point, I would encourage ostracizing Molyneux; it’s clear from an analysis of the first quarter of the book that he’s wildly philosophically incompetent. But I disagree with that tactic on both a pedagogical and moral basis; those who argue poorly can be taught to argue better, and we (as a community) are best served if those who can argue poorly are taught to argue better, so philosophy teachers ought to do that.
Perhaps the only reason many of us are aware of Molyneux is his insistence that various things are not arguments. This continues in the book. He notes “Taxation is not an argument, because you go to jail if you don’t pay.” (Loc 568) This is obvious, though it’s also obvious that dinner isn’t an argument, because potatoes aren’t premises. His positive account of what an argument is involves mixed metaphors and strange assertions. In his introduction, he insists that valid arguments with false premises are aren’t arguments, which is weird.
For the purpose of making this conventional, I just want to use a minimal working definition of argument. An argument is a series of premises adherent to some rules of inference that produces a conclusion. Molyneux often refers to “The Argument,” and so I will set aside that term for his definition.
The following is what I take to be a primary argument of Molyneux’s book, that “The Argument” is a primary condition for civilization.”
The Argument is the primary non-violent means of changing behaviors or attitudes. (Loc 530, 538)
1.1 A “non-violent” change in behavior or attitude is one where an agent changes their behavior without the threat of force.
Civilization requires principally non-violent means of changing behaviors or attitudes.
Therefore, the Argument is a primary condition for civilization.
One can see why (1) is compelling; it certainly happens in everyday life. If you and I disagree about whether we should go to the movies or the park, then I can make an argument in the hopes of influencing your behavior. Of course, there are lots of other ways one can non-violently compel behavior; Molyneux should know some of them, as their basic stock-and-trade of libertarian thought. For example, instead of presenting arguments, we can make an exchange, agreement, or promise that will lead to satisfying both our preferences.
With this in mind, it looks like (1) is just false on a strict understanding of “primary.” There are lots of other ways to modify behavior. Hobbes, Locke, and Rousseau all expound ways in which humans can develop contractual relationships by presenting their own interests and creating conditions agreeable to all signers. Rawls expands this to include even possible signers. This is a fairly straightforward view in modern political philosophy, and on this very broad sense of contractarian, it seems as though nearly all modern philosophers about it.
But Molyneux’s argument gets worse, because it turns out (2) is also false. Molyneux notes that “Freedom is freedom from force,” and maintains that civilization has to involve non-coercive engagement. The problem is, even if we take “non-violence” to be the absence of force, it is still hard to say what he means by “force.” After all, governments use force to address dissenters, and Molyneux hopes to avoid force by advancing a successful argument (Loc 1292).
The problem is that it is not clear at all when force is being exercised. Is “force” only the actual or possible commission of violence by another agent? Or can conditions “force” an individual to do something?
Suppose a cancer patient goes to see a doctor; the doctor has access to the chemotherapy drugs that the patient needs for treatment. Does the patient’s position force them to make a purchase? The patient isn’t in the same position I am when I look at buying a new pair of jeans; she can’t simply walk away. If she has access to a different doctor willing to provide the treatment at a lower cost, then there’s market possibility, but she’s dependent on the presence of alternatives and the good-faith pricing practices of sellers, precisely because she isn’t “free” to abstain. If she doesn’t buy the drugs, then she’ll die.
If contexts can be coercive, then plenty of instances of argument are coercive. Philosophy classes are coercive, though the stakes are a lot lower than relationship between a patient with an acute condition and doctor. Students who attend my ethics class require the course to graduate; they need to graduate to go into their chosen profession. Could they leave my class? Sure. But the quality of life for those students would suffer significantly, both in terms of career prospects and their own personal preference to pursue a particular sort of career. Is the mere contextual feature a matter of force? Sure. It isn’t violent, but it influences students to take some arguments more seriously than others; as a teacher, I have a responsibility to make that force work for the best interests of the student, but it is silly to deny that it is there.
Molyneux doesn’t address whether these instances count as force, but he does say some very confused things about force. He insistence that the primary mechanism of managing those who do not engage in civilized society is ostracism, but it’s not. It is force, whether forced imprisonment or some other severe consequence. Social contracts are enforced; social norms are enforced; an ostracism itself is a form of force. It isn’t as violent as hitting someone over the head with a club, but as an ex-Mormon about the impact of disfellowship on their life and it becomes clear that it is a powerful, and coercive, force.
As this line of analysis continues, Molyneux’s account looks even worse. Claims about the sensitivity of standards for argument to social factors, the role of concepts, and the like come back to bite him. But, up to this point, it’s clear that Molyneux’s argument fails, since both premises are false.
Molyneux’s argument fails (in part) because it can’t even meet his own basic standards of clarity of definitions. He writes, “Almost all human conflicts result from a lack of clarity in definitions.” (Loc 1173) As absurd as that evaluation of human conflicts is, the notion that definitions matter is right. Whether screwing up definitions of truth (Molyeux’s view should be, “Truth is a two-place relation between a proposition and the world.” Introductory level philosophy for objectivists about truth.)
The more interesting point in this discussion is that Molyneux’s argument is not just demonstrably false and a clear conceptual failure, but that if one takes the argumentative conventions seriously, they should reject the various libertarian claims he makes in the book. Some arguments are bad because they make rudimentary philosophical mistakes; that’s surely true. But Molyneux’s is especially bad because it intends to advance objectionable (and even laughable) theses that includes demonstrably false claims about interpersonal relationships and how they manifest justice in a society.
Much of the book rehearses a familiar strand of libertarian thinking; there are some arguments in the libertarian literature that are philosophically interesting, but Molyneux doesn’t touch on them.
Molyneux maintains that in a “free society,” i.e. one that conforms to the moral values he advocates, the success of individuals is responsive to the quality of their work. If Timmy works hard and Steve doesn’t, then Timmy will be more successful than Steve, but this argument trades on a few assumptions that are demonstrably false. These assumptions are that, in such a society, (1) individuals start off roughly on an equal playing field and (2) individuals’ improvement on the playing field is the result of hard work.
I want to focus on condition (2), because it’s failure is the most destructive to general libertarian arguments like Molyneux’s, but (1) is interesting. It is transparently false that a libertarian society entails that individuals will start in a roughly equal position. Over generations, in a state where individuals can accumulate wealth and pass it on within their family, some individuals are going to be positioned disparately in ways for which they are not morally responsible. If Timmy is born to parents with no educational background, low income and financial security, or other issues (e.g. alcoholism in the home; abuse; etc.) then he is in a demonstrably worse starting position than Steve, born to an upper middle class family, much less one of the Trump or Kennedy children.
Wealth inequality is a fact of any libertarian society, but also of most idealized liberal democratic societies; extremely disparate wealth inequality is only a feature of the former, which makes his own libertarian position untenable. Liberal democracy, at least in formulations offered by Rawls, aspires to putting Timmy and Steve and Baron Trump on a par (though not necessary an equal position), so that it’s plausible that their success is a result of hard work, rather than the circumstances of their birth.
The further problem, and the more egregious error, is that (2) doesn’t track either. A libertarian should be aware of this, but it is common to ignore. Suppose Timmy and Steve (starting on an equal playing field) go in to work for two different companies; Timmy works harder than Steve, the quality of his work is demonstrably better, but either due to errors of management or demand for the product, it goes out of business and Timmy is unemployed. Timmy is not at fault, but is now in a far worse position in his career than Steve, who has a secure position at a company that has invested in Steve’s jobs skills.
This is one style of case where someone’s economic circumstance is not responsive to hard work, but rather to external circumstances; we see this practically in contemporary coal mining, auto-manufacturing, and other fields that has seen enormous job loss through no fault of the workers. The workers then unjustly carry the economic burden of the company’s failure. This isn’t a bug that the libertarian position is equipped to result.
But Molyneux’s position gets even worse. Molyneux insists that a labor market responds to the quality of work, but anyone who has spent significant time watching folks get hired and fired knows that this isn’t true either. Labor markets are responsive to the traits that managers value. If Steve has some other trait that Timmy doesn’t (perhaps he’s a relative; or more sociable; or holds similar ideological views), and prospective employers value that feature, they may overlook the fact that Steve doesn’t work as hard or as well as Timmy. Circumstantial factors matter to hiring and firing.
If we hold the view that working hard and well ought to secure someone at least basic financial safety, a point Molyneux insists it will if the market is allowed to flourish, then the above demonstration shows (contra Molyneux) that a libertarian society will not, and cannot, address this.
A 45-year-old coal miner who loses his job when the mine closes down due to shifts to alternative energy, or an autoworker who loses his job to automation, isn’t faring worse than his friends because he didn’t work as hard. He’s faring worse because market forces he couldn’t have anticipated at 18 have devalued his skills. If libertarianism roughly entails not acting on those market forces and precludes redistribution of resources to compensate for the circumstance’s unfairness to those workers, then it does not provide an answer to this issue. After subjecting you, dear reader, to so much Molyneux, perhaps it is worthwhile to turn to a decent philosopher more attuned to social obligations to fairness and justice. In his seminal discussion, “Justice as Fairness,” Rawls gives a note sympathetic to much of Molyneux’s account of rational deliberation, but without the glaring philosophical errors of general reasoning and political philosophy outlined above. The result is an articulation of the basis of Rawls’s political liberalism:
“The fair terms of social cooperation are conceived as agreed to by those engaged in it, that is, by free and equal persons as citizens who are born into the society in which they lead their lives. But their agreement, like any other valid agreement, must be entered into under appropriate conditions. In particular, these conditions must situate free and equal persons fairly and must not allow some persons greater bargaining advantages than others. Further, threats of force and coercion, deception and fraud, and so on, must be excluded.” 1985. “Justice as Fairness” Philosophy and Public Affairs 14.3: 235
Note: Since I read this on Kindle, all of the citations refer to kindle locations in the book, rather than page numbers.