It's not offensive yet ;-).
First, your conclusion that putting in backwards compatible support two CPU architectures ago was "forgotten". DEC's (note that I spelt it DEC, not Digital) style was to not try to implement the past but to provide a seamless move of bits from old architecture to new architecture.
Second, I would expect trying to "emulate" VAX to be more expensive. To make a VAX set of instructions cheaper would defeat the long-range goal of converting the support and maintenance to the new architecture. One of Wang's problems was they got themselves stuck in supporting everything they ever did. Not only is this impossible but it sucks money big time for no benefit.
Third, I asked about who the hardware architect was because I can't remember. If I knew who it was, I could probably make an educated guess about how things happened during the design and first breadboard phase of the project.
I guess my main critcism would have to be that they had the wrong design goals. Had I been running the group and they came to me with the choice you...
Fourth, I did know the program manager of Alpha. He would not have ignored FORTRAN, COBOL, nor any other language.
Fifth, JMF did a lot of the Alpha work and I don't remember him muttering about any of this. Note that I said "remember"; my recall is extremely faulty.
keith easier instructions to I don't think INTEL made the same trade-off at all, they had a far better process and about 2 orders of magnitude more...
Subtract a hundred and four for e-mail.