Shoop Shoop: Turing Failed His Own Test How Can Anyone Expect Anyone Else To Do Better?

If I'd thought of it as I was writing yesterday's post dealing with the materialist faith that computers do a version of what we do when we think and act, I would have said that you'd hear some echo of that nonsense pushed in the media within the next day or two.   It wasn't a day before NPR delivered on that unmade, perhaps unnecessary prediction.   Their film reviewer, Kenneth Turan, did a short piece on a movie called Ex Machina, which revolves around a computer- well a robot, the fabled "Turing Test" and the heart and mind of a young geek who has to decided how he is supposed to take the sexy computer gal his eccentric jillionaire boss introduces him to.  In other words, a 428,569th variation on Pygmalion and Galatea presumably without an Aphrodite to make it real. Perhaps the math god does that magic.

I am sure you all know what the "Turing Test" is supposed to be, if someone has a computer they claim can think it is to be tested by asking it questions.  If the person asking the questions and getting the answers can't tell the difference between the computer and what they think a person would answer to those questions, then the machine is to be declared a thinker.   Though Turing's actual concept of his test was considerably more sophisticated, that's what it is for most of us thanks to the vulgar popularization of science, what science is for most of us.   I doubt most of the guys who throw the term around would even read his paper, but here it is.  Maybe I'll go through it sometime but on a quick reading of it, Turning's argument rests on some rather large leaps of faith and unfounded assumptions.  He seems to take for granted the ability of people to discern authenticity on the basis of those kinds of judgments which is absurd.  His test presumes a power of discernment in people that simply isn't reliably there.  The test fails without it.

I'm old enough so that for several months in the mid-1960s I was first charmed, then rapidly afflicted by the constant background music of  "The  Shoop Shoop Song". The song where the fine pop singer Betty Everett asks how she can tell if her boyfriend really loves her.  As the backup singer gals propose a number of tests, she declares that "If you want to know if he loves you so, it's in his kiss" reiterated for emphasis, "if it's love if it really is, it's there in his kiss".  Now you know what's going to be going through my head all day.

The disconfirmation of that theory was obvious all over teenage America in the form of girls and boys finding out that despite having been convinced of finding true love by lots and lots of that kind of testing, they later discovered that the test was totally bogus.  The results in the form of breakups, tearful, angry and,  after the Hollywood example of the times, generally quite histrionic.  And, superior to  the Turning Test, the evidence for judgement of the entire Shoop protocol was far richer than just the "kiss test" you had to "kiss him and squeeze him tight to find out what you want to know".  In the context of real life application the decision would have included answers to questions, quite often the answers given to those questions initially believed with all the heart, only later to be falsified by other evidence.  Clearly teenage girls of the early 1960s required a higher level of evidence than that provided by the Turing Test.  And look at how well it worked out for so many of them in making important decisions.

As I recall, science majors, even those in math and the burgeoning field of computer science were susceptible to being deceived by the far richer range of questioning and testing of the hypothesis that the genuine nature of their boy or girlfriend was reliably revealed to them.   I never did a calculation to see if their saturation in the thinking of science made them more successful in discerning sincerity in their lovers than those of us in the humanities.   I can remember loud, noisy angry breakups and accusations of insincerity and betrayal in sci-majors. But, then,  I am sure that it wasn't science that informed my caution in trusting my feelings in love or my conclusion that wishful thinking was often a dangerously deceptive component in coming to such conclusions.

The idea that you could determine the genuine status of what was going on in a machine on the basis of your ability to be deceived by its output is contradicted by the most obvious and widespread experiences of humans interacting with other humans.   If we're so unsuccessful in determining if the expressed thoughts of other people are sincere, if we're so able to be deceived about those, completely misjudging the people who tell us that, there is no reason to think testing a machine will yield reliable results. THE TURING TEST PRESUPPOSES THE ACCURACY OF JUDGING PEOPLE ON THAT BASIS [I add because apparently fans of the Turing Test don't realize that ability is an absolute prerequisite for it to work.] The idea that you can use it in science is about as silly as the idea that economic choices are made on the basis of "reason" which is the oh, so, hoped for thinking in many of the same folks who push this kind of nonsense.

Alan Turing should have known better but it was just the same kind of wishful thinking that got him in trouble with the much younger guy he took home, who ripped him off and who Turing so naively and stupidly reported to the police, with such tragic results for himself.  I don't know much about his dating history before then but anyone with a brain should have figured out what he so obviously hadn't about judging other peoples' thinking, both the kid he picked up and -so stunningly cluelessly for a gay man of his age -  the cops in the 1950s.  I wonder if he bought a similar message encouraging wishful thinking from a silly love song from his youth.

--------------

I have thought more about it and looked at some things and I am more convinced than ever that any materialist model of the brain can't work unless precognition or some other psychic faculty IN THE FORM OF NON-MATERIAL TRANSFER OF THOUGHTS is part of it.  There is simply no other way for the brain to correctly construct the material basis of ideas and thinking unless the idea was already present to inform it of how to do that.   The atheists, the "skeptics" the "brain-only" guys can either accept that their idea requires that precognition or, perhaps, telepathy or clairvoyance must be real or their model fails, completely.  It doesn't work without those.  They can't keep their faith in thoughts as material substance while denying the only thing that makes it work.

Someone, tell me how you believe you could get a brain to construct the right idea or even wrong ideas, which would have to be as much the material construct of the brain, without having something to go on in the form of the idea that it was supposed to construct already being present in it.  I don't think you can do it.

Update:  Well, I don't know about being smarter than Alan Turing but I wasn't stupid enough to a. pick up a guy I didn't know who ripped me off, b. make a complaint about the guy ripping me off to cops who would arrest me for sex that was illegal at the time (I'm old enough to have direct experience of that past) and, c. think that a silly party game where people tried to figure out if a woman they were asking questions of was either a real woman or a female impersonator was a basis for deciding if a computer could think in the same way a person could.  Even really smart guys can be really stupid about things they want to be true.  You're wrong about what I said, which isn't what you want me to have said, but you definitely ain't no Alan Turing.