I advocated for anonymous student evaluations back in the early '70s. What we have found is that anonymity allows irresponsible evaluations. Anyone who has been divorced and had anonymous child abuse complaints made understands this process as well.
Confidentiality, which would mean the professor doesn't know who wrote the evaluation, but someone does, gives a much better balance of rights.
I agree with what you're saying and I have some thoughts.
Evaluation isn't merely speech, but requires discernment and judgment. If you're going to make a qualitative judgment then it is more like voting and I agree that should not be completely anonymous. The way it should be is that the professor in question shouldn't know which student voted which way, but the *computer system* itself could know. Access controls could be put in place so that the professor does not know. In fact no specific person would have to know until there is a dispute and a valid reason for access, but the *computer system* should always know.
For example if you look at it like pulling scrambled identities out of a hat and seeing which way each one voted, I think this is completely fair. You could get the polling information, and it could be pseud-anonymous in that only the computer would know, but there would also be accountability because if something goes wrong it could be unscrambled rather easily. I don't think the professor needs to know how each specific student voted. I don't think a human being has to constantly monitor it, but only act as an arbitrator. In a dispute the arbitrator should be given access to the identities.
I understand also that complaints should not be truly anonymous. A complaint is once again something which is a qualitative judgment which has a cost/risk to the individual being judged. I do think that the person being judged doesn't always have to know they are being evaluated, but I also think that if they are evaluated the people who evaluated them must be known by a third party. So I don't advocate a black box, I advocate a grey box where you can put your vote into it and have the ability to voice your concerns. without your boss, professor, or whomever knowing exactly who you are. This does not mean the computer system should not know who you are. It should always know who everyone is.
To reverse it for example let's say the professor really is corrupt and students need a way to report this corruption, there should be an pseudo-anonymous channel to allow them to report this. In this case they would be considered an anonymous-source. However this can be made pseudo-anonymous by attaching a digital signature, so that now they have an alias and are not truly anonymous but pseudo-anonymous and their reputation can now be mapped. If they give bogus reviews under their pseudo-identity then the computer would know this and since they have to be a legit student to have a pseudo-identity the computer could revoke their ability to give reviews if it is found out that they are abusing it. The professor does not have to know which student said what, the professor could dispute that the student giving that bad review is biased and the computer system can look up a history of reviews to find out if the student is biased or not.
It's really no different from Ebay or Amazon. The reviewer should be reviewed by the system and held accountable, and the people being reviewed should be reviewed by the system and held accountable. Unreliable pseudo-anonymous sources will be flagged and labeled as such. Reliable pseudo-anonymous sources will be flagged and labeled as such.
If I'm the professor being reviewed I just want to know I'm reviewed by people who are honest, competent, reliable and as unbiased as possible. Anyone can smear me pseudo-anonymously, but it's a lot less likely that people who don't have a history of doing that will choose me out to smear, and if that did happen there would still be patterns which could reveal sub terranean network organization. If I'm a reviewer I don't see why my reviews could not also be pseudo-anonymously peer reviewed.
Thank you for your post. You've made me think.