Dismantling the argument of “child actors become drug addicts”

This will be a very simple and short argument from me because it’s not really a hugely relevant topic and most intelligent people already do understand basic fallacies in arguments. But I still have a need to illustrate how thinking may rot as I keep seeing gullible people jump to conclusions.

This is the argument I will be criticizing:

Wow, this guy is a drug addict? I remember him from “20-year-old movie”. I can’t believe how far he has fallen. But I guess it was inevitable with him being a child actor and all. So many of these cases!

The claim goes that being a child star is so stressful that it leads to drug abuse.

Let’s assume this is about child actors in Hollywood only. We, therefore, exclude singers and other entertainers to simplify my argument. Drug abuse and child acting is not really defined here so I’ll just try to look past that and talk about generals.

To understand such arguments we need to look at the structure of the claims. What’s the claim? What’s the evidence? Does the evidence support the claim?

At first glance, it sounds true. There does seem to be many child actors who become drug addicts and we all can mention a few such examples. And clearly acting for a child may hypothetically be a stressful profession. Most children don’t have jobs or just have menial jobs. So child acting does seem like something extreme that may hypothetically have extreme negative consequences for some people.

I have structured my counter-argument into 4 counter-claims.

1st error — assuming that overall effects can explain what happened in this specific case

I want to get this out of the way as it doesn’t really dismiss the overall claims of it being possible or it being statistically true for the group overall.

Extreme stress may on average cause a higher rate of drug abuse in the group. We know 50% of alcohol abuse is heritable for example. But we can’t just point at a single person and go “This person’s addiction if 50% caused by heritability”. It may be 70% caused by heritability or 30%. The average for the group is just 50%. Still, if we know the overall effect sizes we can assume it is the case for the specific person and treat the person accordingly.

The problem is much bigger than this though. Because the effect sizes themselves are calculated based on terrible subjective criteria. So the argument proposes that something happened to the individual without even showing that it happens at all.

2nd error — estimating effect sizes from anecdotes

Humans have instincts/biases that estimate effect sizes based on anecdotes. So a few personal experiences with certain people or just hearing anecdotes about these people make us assume things about their group overall. It’s a bias we have evolved as we need to learn from mistakes and generalize the little data we observe. But it’s called a bias as it fails systematically compared to logical calculations. Logical reasoning requires more data, better data and more meticulous calculations before reaching conclusions.

The assumption biases reach these conclusions in the argument:

1. Assuming you know how many child actors become drug abusers
2. Assuming you know how many child actors there are all in all
3. Assuming you know how frequent drug abuse is overall in the city
4. Assuming that child actors start out being similar to the group average

Let’s respond to the 1st point. The argument supposed that the person knows how many former actors are drug abusers and that the number is high. The problem with the argument is that it’s based on bad data. It’s not a random sample, but rather a systematically biased collection of data. We all know of former child stars that fail at life because we are curious about seeking out these cases and reading about them as we are very interested in famous people and everything they do. We seek out stories about former actors failing. And writers know this and try to seek these cases out too to sell papers. We don’t read all the stories about thousands of child actors who turned out perfectly fine and got regular 9-to-5 jobs. And even if you do read about such cases they are often not salient enough to be remembered long-term. You remember 10 huge cases of failure that are used as horror stories to warn people. Just like you remember the 5 cases of school shooters who also played PC games. They are warning stories used to scare people into behaving a certain way and enacting certain laws. There are always an anecdote to support any claim.

The 2nd claim is part of the first claim. If there are many child actors who become drug abusers it must imply that you know how many child actors there were all-in-all in the generations your scary examples are taken from. 1000? Does it feel right? How many TV shows and movies come out each year with child actors? What is the real number based on facts and not feelings? Without a calculated number we are just feeling our way to a number. This will just make our biases work overtime.

Let’s go to point 3. The argument proposes to know how big a percentage of the overall population are drug addicts as the child actor cases are compared to this number. Let’s just say 1% of the US population are drug addicts in any given year. So 1 out of 100. How many child actors become drug addicts? If acting has a negative effect we should expect the percentage to be higher. If you can spot the difference without doing slow statistical calculations on the numbers then it must be a fairly large difference. We are not talking about 5% as you probably wouldn’t be able to spot a pattern from such small differences. I assume the group differences must be about minimum 10–20% to be seen with the naked eye.

All of this makes us able to compare small groups to bigger groups. But the most telling point of them all is point 4. Are child actors representative of the large group they belong to? Are they born with similar traits? Do they have average parents? Do they come from the average SES? Because if child actors start out being different out whole argument is bunk right away until we explain away all these differences.

Though we don’t always need to calculate these differences to know they are there. We can for example clearly see in both studies and real life that girls naturally prefer playing with dolls and boys prefer playing with toy cars. But if someone is critical of these claims you still need to find the effect in the numbers and present the numbers.

Okay, so let’s say we have defined the terms “child actor” and “drug abuser”. We look at the percentages and compare them to the city overall. Have we proven our claim? Not even close.

3rd error — assuming an effect

Now let’s look into the stress hypothesis. The whole argument proposes that extreme stress may cause drug abuse later in life. This is actually true. Extreme stress is unhealthy.

Do child actors typically experience enough stress to cause them great long-term harm? At times that may be the case, but not always. We know that plenty of child actors grow up to be relatively healthy and successful adults like Leonardo Dicaprio, Christian Bale, Jake Gyllenhaal, Joaquin Phoenix and Kurt Russell. They most likely have experienced great difficulties in life like all people, but they are focused on work and live successful lives. Though it may seem like all of these people are more crazy than you. When you freak out and thrash your room no one may be looking. When stars freak out everyone is looking and judging.

So in at least some cases the acting stress may not be great or their personalities may just handle the stress better. Or maybe just remaining a popular actor calms the mind? That may be true, but plenty of child actors quit the business and grow up to become healthy adults too. We yet again need to look at the numbers because my counter-anecdotes as just as good as the scary anecdotes. At any rate no clear reasoning is presented in the argument about the effect stress has on child actors. It’s just assumed to be true maybe because we want kids to remain kids and we don’t trust Hollywood so we tend to be critical of both things. I agree with both moral doctrines, but I trust Hollywood more than I trust anecdotal evidence.

From the argument alone we don’t know if there is enough extreme stress in child acting on regular basis to cause the huge negative consequences people are talking about. Just because something is possible doesn’t mean it happens in the specific example you talk about. For all we know child acting may cause this weird group of people to be less likely to become addicts.

4th error — causality

This one is by far the most irritating error here. The error of causality is so often seen in logical arguments that it makes me throw things around and punch computer screens — nearly. First comes the error of calculating stats based on a few salient examples. Then comes the reasoning based on subjective worldviews. Clearly the stress gets to these kids! Right? No, even if we did statistical studies on child actors and later drug addiction and showed that there was a big correlation we still would not have shown a causal effect. Correlation does not imply causation. There are vastly different array of effects causing drug abuse.

Alcohol use disorder is for example about 50% heritable:

AUD is approximately 50% heritable. The multiple genetically informative studies of this syndrome have produced consistent results that support the validity of this heritability estimate …

There is no need to go against Occam’s razor and create more variables than needed to explain an effect. Some people are just more likely to get addicted to certain drugs. This is how simple it can be: “These child actors may have a propensity to seek out acting roles. And this exact propensity is what also makes them seek out drugs too.” There is no need to add 4 extra steps to your calculations. We may as well assume the most simple thing or just say we don’t know. If we find evidence to dispute the basic assumption we can start to explain the effect with more complicated variables. If you have a tendency to complicate matters just say you don’t know much about these things instead. Don’t scare people into enacting laws that may do more harm than good.

Certain personality types do in fact correlate with drug abuse. Different Big Five personality factors correlate with different kinds of drug abuse. D. Nettle, (2006) found that actors also tend to have personalities that differ from the average population. This would explain as least some differences in drug consumption between the population on average and actors. But maybe there is even a sub-category of actors that become big drug abusers. This is the sort of stuff we do studies on. But only after a correlation is found.

Another more complicated explanation than, they are born this way, is to assume that the parents push kids into extremely stressful scenarios and overall are terrible parents. So the parents cause the effect. Here the parenting style that causes the kids to act also harms them overall mentally. This does happen in specific cases, but it also happens that parents are terrible and don’t force kids into acting. We don’t know if actors’ parents are worse than average. Plenty of parents don’t want their kids to go into acting as it’s a field that usually doesn’t pay well if at all. Also, the heritability of personality traits is high. So even if the parents have an influence on kids the biggest influence may still just be the genes the kids inherited. And typically we don’t really find any effects from parents/teachers on kids. So unless they are horrible parents they will not have any influence on their kids on average.

Conclusion

To assume that child actors become drug addicts because of the acting is a big error in judgement. For this conclusion, you must assume several big things about personality, child actors, drug abuse and society. While just relying on anecdotal evidence.

Actors are people we love to read about so if some of them develop a drug abuse we love to read about it and blame something we hate for it. In reality the most simple explanation is that they just are these kind of people and that acting didn’t really influence them much at all. They would have become drug addicts at any rate. They just also got to be child actors and maybe that helped more than it harmed.