Curved in Upon Ourselves

Correlation doesn't equal causation—unless you're in a feedback loop

Earlier this week, I attended a summit at the National Center on Sexual Exploitation on the upcoming Supreme Court case, Free Speech Coalition v. Paxton, which could determine the future of any efforts to regulate children’s access to hardcore pornography. One of the presenters, Lisa Thompson, shared the results of a recent study that demonstrated that teens who regularly watched pornography were more likely to (1) have much worse relationships with their parents, (2) have poorer academic achievement, and (3) show a propensity to acts of sexual harassment or violence. Today, another of our collaborators in this battle, Michael Toscano of the Institute of Family Studies, published an article at the IFS blog documenting a recent survey that showed that frequent porn consumption doubles the risk of feeling depressed or lonely.

When hearing Lisa’s numbers, I couldn’t help hearing the voice of a devil’s advocate (in this case, it really is the devil’s advocate!) in my head: “correlation doesn’t imply causation.” The porn industry will tell us that of course, teens who are lonely and depressed and have bad relationships with their parents are more likely to take refuge in porn, and that those who have a sexually predatory streak will be more apt to want to watch porn too. They might even suggest that lazy, unfocused students are going to be the ones with more time for watching porn anyway. Now of course, none of these retorts place their industry in a very good flattering light—“So what you’re saying is that your product is best suited for depressed, anti-social, predatory drop-outs?”—but at least it gets them off the hook for causing the anti-social behaviors.

In following Jonathan Haidt’s Substack, I’ve noticed a similar theme. For the past couple of years, he’s been playing whack-a-mole with more tech-friendly sociologists who insist that the connections he’s documented between social media use and poor mental health don’t tell us anything about causation—maybe it’s just that otherwise unhappy, unstable people are just more likely to binge on X or Instagram? And indeed, they probably are!

Perhaps, however, the very premises of such arguments are misguided. After all, as the classic meme puts it, “Why not both?” Why can’t we say, for instance, both that people who are feeling down on themselves and down on the world are more likely to turn to pornography or Instagram as a drug to dull their depression, and that the more they do so, the greater their feelings of doubt, despair, and self-loathing become? In fact, any pornography addiction counselor will tell you that this is exactly how the psychology works, as a feedback loop: nobody likes me à I guess I’ll look at porn à what a loser I am—why would anyone like me? à I guess I’ll look at more porn. Note that the loop can start at either point: the emotion can initially trigger the action, or perhaps more commonly for young people, an inadvertent encounter with pornography can help get start the cycle.

Indeed, such feedback loops are perhaps the norm, rather than the exception, when it comes to psychology: negative emotions trigger negative behavior, which triggers more negative emotions, etc. The attempt to neatly distinguish correlation and causation is a healthy impulse in the physical sciences, where much hinges on whether the earthquake caused the avalanche or the avalanche caused the earthquake, but in the realm of the spirit, things are rarely so simple. Indeed, one of the classic texts of New Testament moral theology, Romans 1, makes this explicit:

“For although they knew God, they did not honor him as God or give thanks to him, but they became futile in their thinking, and their foolish hearts were darkened. Claiming to be wise, they became fools, and exchanged the glory of the immortal God for images resembling mortal man and birds and animals and creeping things. Therefore God gave them up in the lusts of their hearts to impurity, to the dishonoring of their bodies among themselves, because they exchanged the truth about God for a lie and worshiped and served the creature rather than the Creator, who is blessed forever! … And since they did not see fit to acknowledge God, God gave them up to a debased mind to do what ought not to be done. Though they know God's righteous decree that those who practice such things deserve to die, they not only do them but give approval to those who practice them.”

Failure to honor God leads to mental darkness, which leads to idolatry, which leads to a debased mind, which leads to corrupt actions, which leads to a disordered moral vision, and so on. Do people behave like beasts because they treat God like a creature, or do they treat God like a creature because they want to behave like beasts? Yes.

Augustine’s famous description of fallen man, taken up also by Luther, captures this circularity: man is incurvatus in se, “curved in upon himself.” To be in sin is to be caught in a negative feedback loop, to be a serpent chasing one’s own tail, unsure how to escape. The scientist’s attempt to distinguish correlation from causation, and to identify which way the causal arrow points in any given correlation, is based on the laudable belief in the reality of efficient causes (to use Aristotle’s term). Things don’t just happen, they happen because something set them in motion. But for Aristotle, efficient causation only makes sense in the light of final causation—that is, it is only because we know that something is moving toward an end or goal, one proper to the kind of thing it is, that we can talk about the beginning and end of a process, and thus speak of causation, rather than mere motion.

In the realm of the spirit, then, clear causal accounts are possible only inasmuch as a person is proceeding toward their true end, their final cause—God. Inasmuch as we have turned away from him, and turned in upon ourselves, we behave like beings without a clear final cause. A young child, caught in the act of some wrongdoing and asked why they did it, is apt to answer, “I don’t know.” Often enough, they aren’t just being evasive, they really don’t know. How could they? For sin is absurdity, unreason; to think we can explain it is to try and make it something other than it is. As we become alienated from our true end, we become trapped in feedback loops in which evil is both cause and effect.

This is not meant to let social scientists off the hook from hard work of identifying causes and effects where they are neatly separable. But it is to insist that when it comes to pathological behaviors, they may not be—and indeed, the more pathological the behavior, the harder to separate.