Before we get started I wanted to make a note that my intention behind this thread is one of constructive discussion. I did not make this thread to discuss how common the issue is, the statistics or research surrounding it, who has it worse, whether or not two drunk people are raping each other, etc. I would suggest that anyone turning it into one is simply ignored. This thread is about problem solving, not blame. It's also about realistic problem solving; teaching people not to do something doesn't stop some people from doing it. Just keep that in mind.
There is a fundamental problem with Rape in our society; our ability to accurately determine the crime. When something is stolen, we can demonstrate the exchange of location and ownership; usually verifiable by recites, the sudden and unexplainable acquisition of an item or finances by parties or demonstrating previous ownership through other means. When someone is murdered; we can provide a body and forensic evidence tying murder weapons to the possession of the assailant. When someone is defrauded, we can point to their lack of mental capacities to understand contractual obligations, we can demonstrate the lack of equivalent exchange of promised services and/or goods. All random examples which change from situation to situation, I certainly can't cover all of them.
But with the crime of Rape, we have the problem of demonstrating consent; which is the critical part of the discussion. Forensic evidence can only show that the two people had contact (sexual or otherwise) with each other, but it has no current ability to determine consent. And of course, forensic evidence becomes less and less affective the long someone waits to report the crime. We attempt to demonstrate consent with the only two things we have; the people's words, and examples of which to determine the veracity of their words. This sometimes involves diving into the two individuals past experiences and decisions which may or may not affect the jury's bias toward them. This has brought us to rape shields laws and legislation that prevents certain questions from being raised.
There have been some suggestions to solving this problem; using trained psychologists and lie detector tests. Neither of these is currently being implemented in standard procedure, and I would wager a guess in saying most people would find some fault in the two methods; psychologists being wrong, psychology not being a hard science, lie detectors being inaccurate based on emotional reaction, the ability to cheat a lie detector, etc.
Solving this problem would address nearly every single concern someone has on the issue; the ability to be taken seriously during reports, insuring victims have solid evidence to use against rapists, insuring the innocent aren't falsely accused, etc. There is no party or position that doesn't win from finding a way to make consent provable (either way) to the authorities or in court.
How would you propose to solve the problem of determining consent?