The cultural miasma is this is the way things are done. You can always buck the prevailing culture and be a black swan, but you have to look at the risks and benefits of swimming that way.
And many more folks than me buy in. Many are heavily invested in perpetuating this model because it gives them a level of perceived (rather than actual) control - if they aren't actually forming the startups, and they need a set of ways to evaluate potential success, than overwork may be one of those obtainable metrics you can check off relatively easily.
This means the sausage grinder may always be a sausage grinder no matter the tinge of the sausage coming out on the other end. I have no idea how to go about fixing this other than to begin not by continuing to bitch about work/life balance but to begin to fix my own equation (intrinsic - what are my goals? How will I reach them?) and others (extrinsic - are there talented groups out there who aren't getting a shot because we aren't telling women they can strut in and say "fuck it, fund me?").
I figured I'd start with a more open solution where women are informed about the risks and returns by those who've been there. I've been fortunate enough (for various reasons) that I've been able to get those meetings and hack the existing network here.
If anyone has more concrete suggestions I'd love to hear them - and better yet see them implemented, pronto. The generic "let's fix the sausage grinder model" comments tend to put my teeth on edge because they lack 1. specificity and 2. actionability and thus 3. utility.
Why we persist with collaborative maintenance of the startup 'mythology' or 'reality' (one person's myth is another person's faith system) is, you're right, the interesting question.
I can only answer that for myself.
Early stage, high-growth, hyperactive tech startups where you're doing a new feature release every 2 weeks and an entirely new product on average once every 12 weeks means, sorry, working near 18 hour days.
This model may not be 'correct,' but it sometimes works. And as a startup entrepreneur, my primary interest is in finding things that work.
It's not like I haven't examined a multiplicity of options. I've read 37signals, and used their method for 2 years prior.
Bootstrapping got me to a partnership with a top US health insurer to build a product within 3 months after launching my company, keynotes at Mayo and Eli Lilly, and consulting for the National Board (which administers the USMLE or 'boards' to docs) so I'm not knocking this as a potential solution - and one to which I might return (have a day job, do the startup lunches, nights, and weekends).
The pathology to overwork is a personal tendency which you accurately diagnose. In my case it's been augmented by choosing a role in an early stage tech startup where that group pathology persists.
I'm not sure fixing the social pathology is a viable solution, but like other pathologies, teaching coping skills and how to manage emergent behaviors may be.