Science and Persuasion

This short essay was originally an answer to a question on Quora asked by some anonymous student. I think it is one of the best things I've written there, and is worth saving over here in a different format.

The question and it's context were this:

Which are the best method to convince a religious person of a fact proven by science? I told my religion teacher that in the first five weeks of development of a fetus, "we are all females" but she denied everything. How can I convince her to believe in science?

PS: I know that the Y chromosome determines the sex of a person and that only the appearance of a male and a female fetus is similar in the first five weeks.

(There's a little confusion in my response and the question because the questioner added the postscript while I was writing. I then edited my answer to account for that. You'll figure it out.)

Here was my response.

Science and Persuasion

Convincing anyone of anything through a single conversation alone is nearly impossible. It almost never happens. It is even rarer when topics get controversial. Do not expect to ever find a "magic bullet" for convincing someone that you are right.

That said, let me offer you three points of advice for scientific persuasion. They won't allow you to win any argument, but I doubt you'll find much success without them.

1. Get your facts right.

First, you must get your facts right: according to Sexual differentiation in humans and everything I ever learned in school, human sex is determined immediately at conception depending on the chromosome carried by the sperm cell. Thus, in a very fundamental way we are not all female until five weeks. That five week number is just how long it takes for fetuses to start developing sexually-differentiated organs.

If your whole argument depends on getting someone to accept the authority of science but you get your facts wrong you both lower your own credibility and the credibility of the authority of science in future arguments.

(I see you have edited your question details to note that you know this bit about chromosomes. If you know this, why do you think that your interpretation of what "female" means is the correct one? Why do you think your teacher's disagreement means she "doesn't believe in science"? This is crucial, and I deal with it in my third point below. In any case, the first point still stands in general.)

2. Use scientific reasoning, not appeals to scientific authority.

Science is not a collection of facts. It is a collection of observations together with interpretations and arguments about those observations. If you aren't able to make this distinction clear you will misrepresent the scientific process and your arguments will be less convincing.

Remember that it takes a long time and a lot of arguing before anything resembling "scientific consensus" is established. Most of the "facts" we learn today in school are the result of decades or centuries of extremely smart people disagreeing about observations until they reach a mutually agreeable interpretation. And even then sometimes new observations arise that invalidate the previous interpretation and the process starts all over.

As an example, let's consider the question of heliocentrism vs. geocentrism. That the planets go around the sun is now something "everyone knows," but it took the brightest minds of the renaissance about 100 years to convince themselves of this "fact." (Not to mention that the ancient Greeks argued about the same thing and couldn't agree.) And it took another 200 years after that to make observations that finally showed conclusively that the heliocentric model is inconsistent. 

Do you know why these scientists came to the conclusion that they did? Do you know the observations and arguments that they made that eventually convinced them? Do you know what the evidence that rules out the heliocentric model is? Could you explain these arguments to someone who doesn't know them? If you try to convince someone of geocentrism without knowing all of this, are you really convincing them of science?

But if you can't explain the reasoning that supports a "scientific fact" then your argument will sound like "Well my priest says your priest is wrong, so you should listen to me." Even if you are ultimately right, you won't be able to make a convincing argument.

3. Make sure you are arguing about facts, and not interpretations of facts.

This is the most difficult point, but also the most crucial. 

This is more difficult than it sounds because you may be arguing for an interpretation and think you're arguing for a fact. In your example, are you trying to convince your teacher that "fetuses do not develop sexually differentiated organs until five weeks" or that "fetuses are all female until five weeks (and therefore biological sex is not fundamental to human identity)"? These are very different claims and require very different kind of arguments! If you are not clear in your own mind what you are arguing, you will have a hard time convincing anyone of anything.

And even if you are certain what you are actually arguing, the other person may not be arguing about the same thing! What we call "facts" are almost always tied up with interpretations.

Imagine that we meet someone who wants to convince you that average African American IQ scores are 15 to 18 points lower than White Americans. My guess is that you, like me, have some immediate reservations about this person. Your first thought may be something like "Okay, but why do you want to convince me of this? Next you're probably going to try to convince me that blacks are inherently inferior to whites or some other racist claim." We may or may not dispute the immediate fact, but there are a whole lot of interpretations that are closely related to the fact that you strongly disagree with, and so hearing the fact puts us on edge. You may even be tempted to argue against the fact so that you don't have to argue over the potentially racist interpretations. If you do that, won't you look like "you don't believe in science?"

This sort of thing happens all the time! I highly recommend reading this blog post/research paper [1]  about how often conversations and surveys that look like they are about science knowledge are actually about religious belief (or the lack thereof). It examines some large scale surveys of science literacy/religious belief, especially about evolution, and concludes that

  "That work shows that there isn't relationship. What people say they “believe” about evolution is a measure of who they are, culturally. It’s not a measure of what they know about what’s known to science."

and

When subjects who are highly science literate but highly religious answer “False” to the NSF Indicator’s Evolution item, their response furnishes no reason to infer that they lack knowledge of the basic elements of the best scientific understanding of evolution.

and

For respondents who are below average in religiosity, a high score in “science literacy” predicts a higher probability of “believing” in “Naturalistic Evolution”—and so does a low score!

That is, when it comes to certain subjects, even if you think you are talking about "science facts" other people are talking about what they believe about religion. Even people who give the scientifically correct answer about these topics may not actually know anything about the science, but are telling you that they have low religiosity. (These, I suppose, are people who were convinced of scientific authority, but not scientific reasoning.)

This is very, very difficult to overcome. If you really want to convince someone of a scientific fact, leaving interpretation alone, you will have to put in tremendous effort to convince them of that. You must convey to them "I am not trying to attack your fundamental values. I recognize that there are many possible interpretations of this data. I really do just want to point out this scientific observation. I respect your values and interpretations, even if I disagree with some of them. If I can get you to agree to this scientific fact, I am not then going to use it against you to make your other beliefs look foolish."

This is hard. Maintaining trust in conversation when interpretations get controversial is one of the hardest tasks there is. It requires trying to really understand your conversational partner and their values. It requires asking lots of questions, and listening sincerely. If you assume that because someone disagrees with you that means they "don't believe in science" you have failed at this, and you will not find success until you can overcome this bias.

But it is possible. People can come together and learn from each other, and we should never stop trying. When we succeed, it is one of the peaks of human achievement. It is worth spending your life cultivating this skill, and I commend you trying to learn it.

Footnotes 

[1] cultural cognition project - Cultural Cognition Blog - How religiosity and science literacy interact: Evolution & science literacy part 2