Ah yes, that study of 99 (predominantly overweight) men from an infertility clinic that definitely shows soy affects sperm count.
As far as I'm aware the evidence just isn't there with respect to soy and cherry picking the single study that shows some potential link is just helping fuel a dietary myth, would be my view.
This is particularly important when correcting misimpressions. If your comment is neutral, the neutral reader can absorb the new information fairly easily. But if you're hostile (as with "ah yes, that, definitely that"), then you're also signaling a pre-existing battle. The neutral reader gets confused by these mixed signals and feels caught in crossfire, which is not a state that's good for learning.
The upside of battle language is that it rallies any readers who are already on your side, but this is not a good move in the HN game. We want curious conversation here, not escalating intensity or repetition of already hardened positions. The value of curious conversation can perhaps be measured by how much the participants, including the silent readers, move in the process.
It may be more educational to contest one that most readers might not realize is problematic (I didn't bat an eye) than one that is obviously flamebait.
I've done that at least 40,000 times. This isn't a linear ordering!
There are many principles, which we're discovering together, around curious conversation. I went into this one in detail because I think it's interesting.
Btw, if you see a post that ought to have been moderated but hasn't been—for example, a "worthless, flame-inducing" one—the likeliest explanation is that we didn't see it. You can help by flagging it or emailing us at hn@ycombinator.com.
Thanks for the reminder, while I might not agree on an individual level where the line between calling out misinformation and partisan-bad-faith-point-scoring lies I think despite the repeated edits I was over the line here.
While it feels good to post, it doesn't really help the quality of discussion.
Yeah, like I get people maybe being wary of soy. There's a lot of mistruths and half understood myths out there. In addition to various contradictory findings from studies [0].
But for a list speaking from a position of expertise it felt quite jarring to see it included as a definitive "don't" and made me wonder about the agenda of GP (soy with respect to fertility is indelibly linked with the far-right "soyboy" meme).
As a chemist the idea that phytoestrogens might have the same impact as estrogen isn't completely buck-wild, but also as a chemist I know that two isomers with the exact same composition can have wildly different effects in the human body. So I'd need to see a lot of solid evidence before treating is as anything more than a myth spread by misinterpreted studies of red-clover consumption in Australian sheep from the 1940s.
> But for a list speaking from a position of expertise it felt quite jarring to see it included as a definitive "don't" and made me wonder about the agenda of GP (soy with respect to fertility is indelibly linked with the far-right "soyboy" meme).
I can see why that would make you wonder. However, a few years ago, the French TV did a take on soy and products containing derivates [0] (French only, unfortunately). It warned about possible effects on the endocrine system.
In relation to your point, there are two notable things:
1. It focused mainly on effects in females (never heard of a "soygirl" meme)
2. French TV is not affiliated with the alt-right (or any right) in any way, shape, or form. It has a reputation of being quite left-leaning. And that's by French standards.
I haven’t found anything compelling about seed oils being awful, but I haven’t found anything indicating that we should actually eat them, though. Have you?
I’m asking sincerely. It seems to me like they’re better than some alternatives if only to reduce saturated fat intake, but generally speaking, if any refined fat is avoided it’s probably to your benefit.
I often wonder if people think seed oils are bad because they eat way too much of it in general. Kind of like, if you take a shot of windex every morning you shouldn’t be surprised when you become ill, but instead people expect it to make them healthy (not actually comparing canola with windex).
Canola oil in particular seems most interesting. Its nutrient composition is more impressive than olive oil.
As for your theory, seed oils only make up 7% of US diet calories. Saturated fat makes up more than twice that. Replacing saturated fat with canola seems like an uncontroversial outcome improvement.
Wow, I was unaware of the 7% figure. And thanks for this article too, it’ll make for good reading later.
I’m certainly in favour of exchanging saturated fat for canola. I believe it was Finland where they made a big push to use canola rather than butter and the health outcomes seemed significant. There were confounding factors, but nonetheless, it was clear that a reduction in saturated fats yielded better life expectancy and lower rates of heart disease. The results were actually in line with what prior research indicated. Finland went from a lowest life expectancy in the world to one of the highest over decades, likely in part because of that transition away from dairy fats.
In that sense I think canola is amazing. I personally avoid both (of course it’s impossible to do so 100% of the time), but I fully support people choosing relatively healthier options.
Anecdotally, while living in Finland for a number of years I found that most households I knew would mix margarine and butter together 50/50, and that was the only form they used outside of cooking and baking.
Perhaps the correlation is the other way around. Men with less testosterone (and therefore a lower sperm count) consume more products with soy, which is common in a vegan diet.
Not saying this is the case, just pointing out where this could be coming from.
How did you arrive at that conclusion? The only studies I could find show no such correlation[0][1].
I remember reading that vegans had higher levels of testosterone when not controlling for body weight, as vegans tend to be less obese than non-vegans. Can't find the source, so pinch of salt here. Makes sense though.
"You know, there are men who don’t want to drink soy milk because they have an irrational fear of phytoestrogens—even though soy does not have feminizing effects on men. Yet they’re perfectly willing to drink cow’s milk, which has actual estrogen estrogens in it! Within one hour of the milk hitting their intestines, estrogen levels go significantly up, and testosterone levels go significantly down."
"If you take men on a high-protein diet—”meat, fish, poultry, egg white[s]”—and switch them to a high-carb diet of “bread, vegetables, fruit, and [sugary junk,]” their cortisol levels drop about a quarter within ten days. At the same time, their testosterone levels shoot up by about the same amount. High-protein diets suppress testosterone. That’s why if you take men eating plant-based diets, and have them start eating meat every day, their testosterone levels go down, and actually some estrogens go up.
That’s why bodybuilders can get such low testosterone levels. It’s not the steroids they’re taking. If you look at natural bodybuilders, who don’t use steroids, 75% drop in testosterone levels in the months leading up to a competition. Testosterone levels cut by more than half; enough to drop a guy into an abnormally low range. It’s ironic that they’re eating protein to look manly on the outside, but it makes them less and less manly on the inside. And, from an obesity standpoint, in general, a drop in testosterone levels may increase the risk of gaining weight—gaining body fat."
This was my understanding as well (I'm a big fan of nutrition facts), but I think I misunderstood the parent. I thought they were saying the opposite – that a vegan diet would lead to lower testosterone.
As far as I'm aware the evidence just isn't there with respect to soy and cherry picking the single study that shows some potential link is just helping fuel a dietary myth, would be my view.