So to admit that an experiment you have performed has failed is not the same as saying "My hypothesis turned out to be incorrect and was disproved." It is saying "I had a hypothesis, then I performed an experiment which turned out to have nothing to do with disproving it." I admit that I had an experimental failure. I misdesigned the experiment and as a result gained no useful information about the subject at hand.
I am a fan of science fiction stories and by the time I was in high school I had read hundreds of Sci-Fi books. Though I read plenty space travel stories, my particular favorites were about time travel. And I had read enough of these stories to realize that whenever a time traveler meets himself in the past, he has to waste lots of valuable time trying to convince himself that he really is a future version of himself. And usually the future self has to recount some memory (often painful or embarrassing) to which the younger self would inevitably reply, "But I never told anyone that." Then the story would get rolling again.
To me it was somewhat unsatisfying. If it's such a painful memory, isn't it possible he muttered it in his sleep? Or, if we allow time travel, couldn't he have gotten therapy in the future giving up that information to lots of people, any of whom could have come back claiming to be himself? I thought to myself, the only way to prevent these problems is to come up with a sign, a password if you will. It has to be easy to remember but have little or no emotional content. And this is the important part, it has to be secret. Ben Franklin said "It is possible for two people to keep a secret, but only if one is dead." So I knew this sign, this password coudn't be shared. I thought one up, one that could be easily encoded into various communication methods, and I didn't tell anyone what it was or write it down or record it in any way. Even my wife can't tell you what it is.
So, since the late seventies I've had this password, and I had to congratulate myself. If it ever became possible to contact myself in the past (and the need arose), I wouldn't have to waste any time convincing myself. I could just produce the sign and we're in business. So, in a sense, at that moment I began an experiment. The hypothesis is "It is possible to communicate with the past."
Of course, the problem with the experiment is that it fails if I never hear from the future. Maybe time machines are invented but I never buy one. Maybe I have the equipment, but I don't see any need to contact my former self. Maybe I die the year before the hypothesis is proved. Or maybe it's just not possible. No way to tell--no light is shed on the hypothesis--the experiment is a failure. Well, to me, this is not that big a deal. The hypothesis is untested now and, while it may remain untested, at least we might eliminate the possibility that the hypothesis is true, the future contacts the past, but the past doesn't believe it has been contacted by the future. Just by making that scenario less likely, it seems the experiment is worth performing. After all, the resources devoted to it are pretty minimal.
But there's another way for the experiment to fail. A way I hadn't considered when I designed it all those years ago. Listen: about 10 years ago I was walking in the beautiful gardens which surround our house. It was morning, just before I had to go to work, and I strolled along the grass path that winds through the lush perennial beds (created by my wife). Suddenly, two very vivid thoughts occurred to me. The first was "Don't drive to work." The second was my secret sign. It startled me because I hadn't thought about that sign for many years.
Now think about this for a minute.
You can see the problem. Let's say some technology is developed which makes it possible to send a message into the past. Unfortunately, we can't send people back, or pieces of paper, or radio signals. All we can do is alter someone's brain patterns. While that may prove extremely useful, it is of course completely impossible to prove that it has occurred. And by its very nature, provides little evidence either for or against our original hypothesis. But now let's say I'm an old man confined to a wheelchair because of a horrible car accident which occurred way back in '98. And the only means which becomes available to contact my former self is this dubious and sneaky method. I know of course that although I have this sign available, this communication will not prove the hypothesis--my experiment will fail (or rather continue to be unsuccessful). But so what? Who cares about the experiment if I could change my history, escape the wheelchair forever? Would I grab that opportunity? You bet I would.
So there I was in 1998, with these thoughts and I realized no matter what I did that morning, the experiment had failed (well, continues without resolution). If I jumped into my car, crashed into a tree, became a quadrapalegic, would that prove anything? No. My thoughts about just such an accident might have subconciously affected my driving. If I jumped into my car and made an uneventful commute, what does that mean? Nothing. Maybe I was driving more carefully due to the warning. If I didn't go? Well, you can see that doesn't help with the hypothesis either.
And maybe it wasn't a horrible accident, maybe I was going to meet someone whom I would, I don't know, go into business with. That business might eventually fail causing much suffering. Or I might hear a news story on the radio which makes me join an organization where I meet a woman who falls in love with me causing my wife to kill her with a meat cleaver sending her to jail leaving me alone and miserable. Or, or... well, you get the idea.
As it turned out I did go to work--after all, I'm not a complete fool. But, not being a complete fool, I went a little later than I had intended to. And nothing bad happened. Except my experiment was a failure.
What would you have done?