It seems to me if you reverse the statement it makes more sense. Men who beat their women need to engage in more "mate retention behaviors", and are probably more suspicious and prone to being controlling.
Yeah. I think if the researchers looked at flower-giving while controlling for those post-violence displays, the correlation likely wouldn't be significant. My dad sent flowers twice a year--birthdays and anniversaries--which is twice a year more than most guys, and he never hit Mom.
I think the vigilance is a far more important predictive cue. A guy who has obsessive thoughts about his partner being unfaithful, whether or not she is, is a hell of a lot more likely to a) call her compulsively and b) hit her. I'd even say that emotional manipulation is probably a somewhat distant second, because there are so many men who use emotional manipulation without violence, but there aren't so many men who are vigilant but not violent. I wish I could see that study.
I always thought sending flowers was a bit suspicious. At least the "flowers as apology."
I agree about that paragraph. Why do they always have to bring evolution into these things?? It's so dumb. They should have concluded "these men are clearly jerks and should be avoided" instead of trying to justify their actions with some weak genetic advantage mishmash.
Yeah...and it's not so evolutionarily significant when it's an aberrant, socially deviant behavior. Being somewhat reasonable with the ladies seems to have been more advantageous.
I don't know if you could call it aberrant when it's as common as it is and always has been. Quite frankly, it's so common it'd be an oversight not to look at how it works as a reproductive strategy. I just think they're ascribing a causative link to what is a fairly innocuous and non-sinister behavior. I bring girls thinks because I like them, not because I intend to beat them later.
I think it's primarily psycho-social dynamics that explain the prevalence of relationship violence by men against women in historically male-dominant societies. There have been societies that have less segregated or unequal gender relations, and their relationship dynamics and rates of violence are not like ours.
I agree with you on the rest, though. I don't think receiving gifts should be considered a risk factor for relationship violence. In a dangerous situation, there are going to be more pertinent warning signs.
I kinda like getting flowers occasionally - along with other things. I think it also gives them a little chance to show creativity beyond the usual roses - however, they usually fail.
Comments 14
Reply
Reply
I think the vigilance is a far more important predictive cue. A guy who has obsessive thoughts about his partner being unfaithful, whether or not she is, is a hell of a lot more likely to a) call her compulsively and b) hit her. I'd even say that emotional manipulation is probably a somewhat distant second, because there are so many men who use emotional manipulation without violence, but there aren't so many men who are vigilant but not violent. I wish I could see that study.
Evolutionary psychology: mostly b.s.
Reply
I agree about that paragraph. Why do they always have to bring evolution into these things?? It's so dumb. They should have concluded "these men are clearly jerks and should be avoided" instead of trying to justify their actions with some weak genetic advantage mishmash.
Reply
Reply
Reply
I agree with you on the rest, though. I don't think receiving gifts should be considered a risk factor for relationship violence. In a dangerous situation, there are going to be more pertinent warning signs.
Reply
Reply
Leave a comment