Jack Maden, Should Parenting Require a License?, at "Philosophy Break" discusses Hugh LaFolette's 1980 argument that the answer is Yes. I've LaFolette's argument, or arguments that are clearly descended from it, popping up more often recently, so perhaps it's worth looking at why it is, and has always been, a worthless argument.
LaFolette doesn't make things easy for himself in "Licensing Parents", in that he intends his argument to be for the most extreme claim -- not that it would be reasonable sometimes to license parents but that everyone should in fact be required to get a license to be a parent. To get this conlusion he attempts to characterize the kinds of situations in which society requires licenses, e.g., "We require drivers to be licensed because driving an auto is an activity which is potentially harmful to others, safe performance of the activity requires a certain competence, and we have a moderately reliable procedure for determining that competence" (p. 183). Similar things, LaFolette suggests, can be said of doctors, lawyers, and the like. It is in fact a dubious claim that we license lawyers because of their potentially harmful activities rather than (e.g.) because governments wanted order on the matter of who could argue before the bar. Likewise, we know that states didn't start requiring licenses to determine competence, because the first driver's licenses didn't have a driving test. The driving exam is actually an artifact of a completely different thing -- the first driving exams were not for driving but for being a professional chauffeur, and states eventually (in some cases swiftly and in some slowly) started extending the same requirement to drivers in general. It is likewise very unclear that medical licenses were developed, or have primarily been used, in order to assess competence for the preventing of potential harm; doctors started advocating medical licensing to reduce competition and push newcomers with easily obtained medical degrees out of the market. Perhaps you could argue that, whatever the origins of these licensing systems, they are maintained to prevent potential harms, but it is more difficult, I think, when you look at some of the things that are licensed; in many states, for instance, hairstylists have remarkably strict licensing requirements. The point of saying this is that the primary justification for licensing seems never to be merely avoiding harm but the bare fact of extending regulatory powers, which might be done for any number of reasons.
LaFolette argues that it is desirable to regulate any field which meets these three conditions: potential harm, need for a certain level of competence, and moderately reliable procedure for determining that competence. He doesn't consider the many, many activities that meet these criteria for which we don't require licenses -- I teach at a college; that has requirements, but college professorship isn't a licensed profession despite meeting all three conditions. Instead, the handling of the competence issue is done by the ordinary functioning of the job market, and the potential for harm is mostly just swept aside on the general principle that everyone deserves an education. Whether or not college professors have any training to teach at all varies across the board, and most don't, or not much. Nobody, however, takes Ted Kaczynski as an argument that the academic profession should be restricted by licensing -- or, at least, nobody has yet.
In any case, LaFolette argues that parenting meets all three requriements. Parenting allows for child abuse, which is quite harmful both to the child and society. (He ignores the fact, known even in 1980, that parents are the people least likely to abuse their children, but I assume that he would say that the tiny percentage of parents who abuse their children cause disproportionate harm. But I also suspect that he would argue that most parenting methods -- spanking but also many of the psychological methods that were beginning to be used and have become very common today -- are either physically or emotionally abusive, or both.) He takes it to be obvious that there is a minimal competence required for raising children. (He never, however, actually says what it is, despite the fact that this seems relevant to the argument. Walking across the street is a potentially harmful activity, one that often causes wrecks, and it does require a minimal comptence, like being able to read signs and lights, and it would be pretty easy to test for this competence, but we don't have pedestrian licenses. Arguably one reason is that people think that the minimal comptence is so minimal that it's not worth the trouble of testing. It matters, if you are arguing for licensing parenting, whether you think 'minimal competence' for parenting is ordinary human decency or advanced psychological training, or something else.)
LaFolette then goes on to consider two kinds of objections against his argument to this point. Objection 1: Licensing is not theoretically desirable because, regardless of the potential for harm or the need for minimal competence, people have rights that take precedence. LaFolette's response to this seems to me to be mere handwaving; he argues that these rights are "not without limitations" (p. 186). He also gives a completely inadequate list of possible interpretations of "right to have children", and doesn't consider the possibility that people might have a right not to have longstanding customs and usages be interfered with without necessity, or that they might have a right not to be subject to regulation in matters that have traditionally not been regulated when they have not themselves shown any incapacity or turpitude. That is to say, he doesn't consider any of the arguments that individuals doing as they think best, not intervention by the state, is the reasonable default for a free society. The statist tendency of LaFolette's argument is never addressed; he simply treats it as obvious that (1) the state should at least usually get involved in anything that can potentially cause harm and requires a minimal level of competence, and likewise that (2) there are no limitations to the state's authority to require licensing except assessment of potential harm and feasibility, and (3) that state intervention in such matters does not have to be independently justified. When he does address worries about intervention, he shifts the subject. Instead of considering the intrusiveness of imposing a licensing system, he argues that there would be few further intrusions for those who have received licenses. Given how onerous the requirements of some licensing systems can get when it comes to keeping your license, this is, like most of LaFolette's naive claims about licensing, dubious, but even so, it's not the right question. The worry that matters is the intrusiveness of building an entire bureaucracy regulating whether people will be even allowed to do something, and the perpetual danger that such systems will be captured to further vested interests. This is not trivial. Even when people support licensing they usually hate having to deal with the inevitable bureaucracy and paperwork. Bureauracies like licensing systems tend to be inflexible, unintuitive, time-consuming, expensive, and, in short, burdensome, even when justified.
Obection 2: Licensing, even if theoretically desirable, is not practically implementable. This could be because (1) we do not have an adequate criterion of what counts as good parenting; or (2) there is no reliable way to know beforehand what features a person will have who will abuse their children; or (3) administration would unintentioanlly misuse any test that could be developed; or (4) administration would intentionally abuse the test; or (5) we could never "adequately, reasonably, and fairly" (p. 193) enforce such a regime. LaFolette's answer to (1) is that we only need a criterion of what counts as very bad parenting; his answer to (2) is what looks to me like a very expensive and complicated research program (one that would have to be ongoing) whose results can't actually be anticipated. LaFolette's answer to (3) I find particularly worrisome; he argues that we shouldn't worry about unintentional misuses unless we have reason to think they would be more common than in other licensing systems. This is, first of all, not the way to design a properly functioning administrative system; you need not dismissal but specific countermeasures to prevent unintentional misuse. Second, unintentional misuses are one of the things that can give you a reason to think that a licensing system should be scrapped, and every functional licensing system continually worries about this. Again, LaFolette assumes that bureaucracy does not require close vigilance but is somehow the natural default, its failures just foibles until proven more serious. Third, a perpetual worry with licensing systems is that they create moral hazard -- that is, people over-rely on the licensing for assessing the competence, with the result that they are less careful and less protected. That LaFolette doesn't even have an actual answer to this sort of problem is itself a worry. His answer to (4) is even worse, since he says that we shouldn't reject licensing parents unless it can be shown that it is more likely to be abused than other licensing systems. Since licensing systems are -- notoriously -- abused all the time, and repeatedly have to be reformed, we have LaFolette yet again arguing that the failures of the state should be ignored as just the natural and normal course of things, and treating its many and well known abuses as barely even worthy of serious consideration, despite the fact that he extends parents -- a much more respectable and highly regarded population of agents -- no such generosity.
Nor does his answer to (5) give us any hope of anything better. He admits that there could be difficulties, but thinks these can be surmounted. His example of how you might do so: "We might not punish parents at all--we might just remove the children and put them up for adoption" (p. 193). This, I think, summarizes in a sentence all that Hugh LaFolette, at least in 1980, did not understand about parenting, even bad parenting.
The most interesting (and least objectionable) part of LaFolette's argument is an analogy he tries to draw with adoption, in which we do in fact impose prior standards on who can become a parent. But while LaFolette does a good job of addressing some objections to this analogy that are not quite adequate, he misses the fact that we do this because adoption, however important it may be, is an artificial legal construct that exists for legal convenience, and thus legislators can make the requirements for it whatever they please. The high standards, which LaFolette rightly notes are much higher than you would expect for a licensing system, are in part because making adoption too easy would cause all sorts of legal problems and in part to address known failure points specific to legal adoption (e.g., people adopting children to make them sex slaves, a rare but recurring problem even under our current system).
One interesting thing that LaFolette says, which I have increasingly seen, is that in his view the reason people resist licensing parents because they see parents as having a "natural sovereignty over their children" (p. 196). (The way this is often put today is accusations that parents see themselves as "owning" their children.) He regards this as an "abhorrent view", but his reasons for this is are quite vague -- parents who hold this view "may well" mistreat their children and even if they treat their children well, wouldn't be treating their children as deserving good treatment but only treating them well "because they want to" and the view in any case is inconsistent with raising children to be adults. None of this is given any backing, and one suspects, given what LaFolette keeps implying about state power, that most of this is projection -- he doesn't think parents should be regarded as having "natural sovereignty" over their children because he doesn't really regard "sovereignty" as involving any responsibility. Just as the sovereign state should in his view license parenting because the state finds it "theoretically desirable" and can, he assumes that parents will do whatever they find desirable and can do. In reality, parents are the adults who are least likely to treat their children in these ways, by all the evidence we have; they are the adults who consistently show themselves to be most active in protecting the rights of their children; and by the millions they raise children to be excellent adults. LaFolette takes the "natural sovereignty" view to be common, but there is remarkably little evidence of any of the consequences he says that it brings with it.
None of this argument is any good. But there is an irony here, as for a while there used to be a general parenting license -- it was called a marriage license. The licensing system was never perfectly implemented in the way LaFolette seems to imagine his proposal would -- from practical necessity, alternative routes like common law marriage kept having to be recognized, and there were all sorts of complications about how to deal with the inevitable unlicensed children -- bastards, as they were called. The requirements were also clearly more 'minimal' and easier to implement than those LaFolette seems to have in mind. Interestingly, that system, despite an excellent reputation and results, had long started breaking down before LaFolette had penned a word, and its collapse was beginning to accelerate. The relative success of state-licensed marriage at its height might give one reason to think it possible that you could have a successful parent-licensing system of the kind LaFolette wants. But if we couldn't maintain a looser, more flexible, minimal system that people mostly liked, its unclear why we would be able to maintain a stricter, more bureaucratic, higher-requirement system that people don't even want.
Now, of course, this was all in 1980. But in 2010, LaFolette revisited the argument and mostly reaffirmed it. Unfortunately, its prospects as an argument had not improved at all in thirty years; if anything, the difficulties of preventing abuse of licensing systems, for instance, are even more widely recognized than they were. Even worse, it's quite clearly this article revisiting the argument that has led to the immensely dimwitted claims, the modernized version of his "natural sovereignty" view, that parents are generally going around thinking of children as their "property", an accusation that is inconsistent with the evidence we have and which is never backed up by anything that could possibly justify it. The harm that can be caused by an academic article is not always great; but in some cases, like LaFolette's active statism with respect to parenting, it is potentially immense. This sort of quasi-fascist and totalitarian notion that the state has right and authority to intervene everywhere where the ends justify the means is unfortunately surprisingly common among academics; and you don't have to invent a nonexistent multitude of parents treating their children like chattel to explain why ordinary people would push back against it.
*****
Hugh LaFolette, "Licensing Parents", Philosophy and Public Affairs, Vol. 9, No. 2 (1980), pp. 182-197.