Thank you for a great response!
Since you're familiar with the fundamental attribution error I'll let you decide if you can make a strong enough case for it to apply here. My biggest problem with the 'theory' is that it's called an 'error'.
Lets make sure we are talking about the same thing. The fundamental attribution error is a classical cognitive bias in psychology where a person has a tendency to attribute other people's errors on internal attributes, but their own failures on external ones.
If your business fails it is because you are lazy and did not work hard enough. If my business fails it is because the system is rigged against me.
Pea-hens don't have big tail feathers because they don't need to attract a mate; instead they select the mate. Humans tend toward a nearly 1:1 ratio; with a history of male-on-male violence skewing that to men being slightly larger.
So, from this we get that in humans the choices are pretty equal. Both men and women get to choose their mate. That is truer now than it has been in all of history (probably).
If we move toward a system where rich/powerful men get many more women (I can't think of a non-naive/pedantic way around it) then we set society on a collision course with war.
Why would this happen? First off you are assuming that the rich/powerful are all men. There are powerful women too, and one they will take more men. And really, there simply is not that many really powerful people around to make that much of a difference. Donald Trump is not going to end up with 3 million wives. Even if every the rich and powerful man end up with 3 wives each the effect is going to be miniscule.
But most importantly remember, the women get to choose as well. This almost completely negates this effect. If we are looking at it as a purely cynical 'women want rich and powerful men to increase their status' than we see that sharing a rich and powerful man reduces the overall status and power she will receive from him. Sharing a lower upper class man with 30 other women is not going to bring much power or status to her, she is better off with a upper middle class man with only one other wife.
If the top 10% have 90% of the women, that leads 90% of the men for cannon-fodder.
See, as long as women get to choose this is the sort of thing that can never happen. Simply because they will choose to bring other men in to it as well. The Harem version of poly (one man with many wives) is the least popular when women get a choice.
I don't think the basic theory, which is essentially "a person can become addicted to another person", predicts or explains this your argument. It seems that if I'm addicted to coke, meth and my three wives then I've got to make choices... I'm happy to hear more of your thinking though.
Okay, lets look at the two separate things, mutualistic interactions and oxytocin attenuation.
Mutualistic interactions are all about how distinctly different groups interact to enhance each other. The idea is that in monogamous relationships there are things that women bring to the relationship that compliments what men bring. I don't really support the whole 'men are from mars' mentality as I think it is overly reductionist, but lets run with it for this discussion. In poly families you have multiple partners, each that bring something to the table that the others don't, that (at least in a functionally family) are focused on cooperation instead of competition. Since no one man or women can bring everything masculine or feminine to a relationship we can easily see that Mutualistic interactions are improved the more people you can bring together in a familial cooperative group. This is how most of the world works with extended family units living together for this very reason. Poly families tap into this exact same dynamic.
Oxytocin attenuation is all about novel experience. The effects of oxytocin is highly subjective based on circumstance. It strengthens pair bonding up to a point, at which point it starts to weaken it. But adding a third person into the mix more than doubles the amount of time it takes for the pair bond strengthening effect to attenuate. Adding too many people into the mix might keep the pair bonding from happening at all, the research on that is really inconclusive at this time. There is a whole lot of questions around this yet. Quite frankly we are not at all sure how oxytocin works. Remember that the entire oxytocin research has been conducted using blood test for oxytocin during pair bonding exercises, but this research is drawn into question by the fact that we know that oxytocin is not capable of crossing the blood-brain barrier. We are simply assuming that blood oxytocin levels is indicative of neuro-oxytocin levels. That assumption might not be justified.
(I'm leaving vasopressin out of this for simplicity, it does not really fit into a conversation about sexuality.)
I'd also like your take on what MoonBeam has to say here:
One person completes another. If one can't do it an infinite number won't either.
It rings with deep sense of truth for me.
This is complex. There is two parts of this.
I don't think another person can complete you. If you are not complete already no one else will ever be able to do it for you. Requiring someone else to fill some hole inside of you will do nothing but make both of you miserable. Find happiness in yourself then share it with others, don't look for others to give their happiness to you.
I also don't think that any one person could ever possibly be everything I need. I am a complex person, with complex needs. I want a great lover. I want a great conversationalist. I want a great thinker. I want a great adventurer. I want someone that is enthusiastic about life and full of energy and optimism that will spur me to try new things and take risks. I want someone that is content to sit quietly on the side of a river and spend a weekend meditating. I want someone that will leave me alone when I am tired of dealing with people.
I can't be all those people to anyone.
It is unfair to ask one person to be all those things for me.