Fixing begins at "someone screwed up". Blaming delves into who screwed up, which is usually not required to fix the problem (since there may actually be no one to blame).
Sometimes that's true. If you have a team of people doing approximately the same thing then knowing which one of them broken something has little value. You just need a process to stop things braking again in the future. But in the case of a disparate group of people who all do different things, you need to know who caused the problem in order to know how to fix it - the "who" and the "what" are the essentially same question.
In the case of Theranos the problem of "what went wrong?" and "who fucked up?" are effectively the same - either the investors didn't do proper due diligence or the founders misled them. You can't fix the problem without knowing the cause. It doesn't really matter whether you couch it in terms of "what" or "who"; knowing one automatically informs you of the other.
It's not hard for an organization to put a system in place to mitigate any particular individual from making a poor decision. Entering into any decision that is high-stakes inherently justifies the cost of the existence or creation of such a system. For example, RFP processes are often required for any spending over a particular threshold on external vendors. Hiring processes often include input from more than one person.
If the investors didn't do their proper due diligence, then they can take steps to be more responsible in the future. If the founders misled them, then it becomes an economic and political issue of how to address and further prevent such problems.
I disagree. A different approach would be to say "Don't know what happened before, but moving forward Ted is accountable for A, Alice is accountable for B, etc". Knowing what happened before serves zero purpose except to shame. All that effectively matters is what happens from this point forward. You put a process in place this time, to monitor what Ted and Alice are doing. Next time, it's your fault when it breaks, because you were forewarned, and you should have known better. Next time, you know what happened, because you put controls in place.
Or just throw someone under the bus and keep repeating the same mistakes.
The argument you're having is exactly why blameless post-mortems are so difficult to insert into existing companies. Shifting from a desire for "justice" to "doing what's best moving forward" is really hard for a lot of people.
There is a segment of the populace who feels that every "wrong" must be "punished", and they generally make life miserable for everyone else. Usually no one will speak up for fear of being punished along with the "guilty".
In the case of Theranos the problem of "what went wrong?" and "who fucked up?" are effectively the same - either the investors didn't do proper due diligence or the founders misled them. You can't fix the problem without knowing the cause. It doesn't really matter whether you couch it in terms of "what" or "who"; knowing one automatically informs you of the other.