It wouldn't be duplicates. A better plan would reuse the same chassis, and have a staggered production. There have, after all, been 16 KH-11 satellites over 30+

It wouldn't be duplicates. A better plan would reuse the same chassis, and have a staggered production. There have, after all, been 16 KH-11 satellites over 30+ years; they surely didn't all use the same technology. Part of the reason for the expense of HST comes from the delay after Challenger. Hitching it to the success of the Shuttle helped make it more expensive. You are right though that it plays to the sunken cost fallacy. Hubble happened because the initial budget was low. Then it ballooned. Congress would likely not have budgeted $10 billion if they knew that was going to be the final cost of Hubble. It's even less likely they would have funded four space telescopes for that price.