It's a good question politely asked, so uncomfortable as the subject makes me I'm going to try. Like I said though, this is an awful ball of wax and it feels dirty to approach it. I'm going to try and tackle it via the question of how one identifies a racist. Let me start with a generic, emotionally less charged example and then try to relate that to the specifics of the Molyneux story.
Suppose two people are talking. The first says, "Minority X children score below the national average on IQ tests. I bet it's because minority X children are more likely to be exposed to lead in tapwater. Our country should have an all-hands on deck effort to solve the lead-in-tapwater problem"
The other responds: "I am not interested in any discussion that begins with 'minority X children score below...'. You are a racist for saying this and I don't listen to policy proposals from racists".
Which of these two is the racist? By the modern definition, it's clearly the second, because he's arguing for a status quo that disadvantages minority X. Whether his heart is in the right place or not, he's contributing to systemic oppression by refusing to act in the face of injustice and therefore a racist. By the classic definition, it's just as clearly the first because he is asserting a difference between races. Whether his heart is in the right place or not, he's deepening racial divisions by this rhetoric and therefore a racist.
Hopefully we can agree on those definitions, if not the rest of this comment is pointless. Regardless, and assuming we can, let's bring this to the topic at hand.
African-Americans are Molyneux's 'Minority X' and child corporal punishment is his 'lead-in-tapwater'. Let's see first how he fares by the modern definition of racism. The wrinkle here might be that lead in tapwater is done to a population while corporal punishment is done by a population. This might seem to make it apples and oranges, but I'd disagree. Anyone who wished could make the case that higher rates of corporal punishment are self-evidently fallout from the abuses of colonialism, and done to the African-American population no less than leaded tapwater.
What's the standard under this definition for determining whether a person is a racist? I would propose the following two tests: if their professed beliefs are sincerely held, and if their proposals would lessen inequality assuming their premise is correct, then that person is not a racist. They may be mistaken in their premises, but in that case it is up to other non-racists to educate them, and there is no use for hostility in that process. By that standard, how does Molyneux fare? I think his life provides ample evidence that his beliefs were sincerely held. He became a public intellectual and subjected himself to ongoing, harsh criticism because he believed it was important people hear his ideas. He spoke often about corporal punishment of children and not once did he vary his message. As to the second test, if his premise is correct that higher rates of corporal punishment lead to worse education outcomes for African Americans, then it's fairly obvious that his proposal of no corporal punishment anywhere would help close the gap. Having listened to him speak, I genuinely believe that if his proposal was accepted and the result was that his children faced tougher competition for jobs and scholarships, he would consider it the best possible outcome and validation of his beliefs. So by that standard, I would definitively judge him 'not racist'.
Of course, Molyneux never made any of these arguments because like most people born before 1982, he would use the classical definition of racism and probably refuse to cede the linguistic territory necessary to make any of the foregoing arguments. The way to not be racist by the classical definition is much simpler and requires no arguments about colonial fallout. One must simply start the argument by saying 'inner city children...' instead of 'minority x children...' and one's thoughts on race are one's own affair. The trouble with that approach is, people born after 1982 immediately start shouting about dog whistles and secret racism, and one finds oneself isolated with people who increasingly egg one on to just name the races in question. I believe this is what happened with Molyneux. When I first encountered him, he seemed to be making a good-faith effort to talk about specific demographics instead of the races over represented in them. By the time I lost interest in his content, I must admit, he was no longer making that effort nor did the bulk of his audience want him to.
So on deeper reflection, I think it was a bit disingenuous of me to judge Molyneux only by one set of standards. By what is probably his own definition of racism - and though I'm conversant in both linguistic systems, the definition I use in private thought - he did commit racism. My only excuse is that I cede linguistic territory as instinctively as Molyneux would defend it, and it didn't occur to me much harm could be done thereby. I'll walk my statement back and say that Molyneux weaponized racism for an agenda that, had it succeeded, would have reduced racial inequality.