Free will, as found in champagne drinkers and robots

hallway_drawing_by_marcychristashawn-d30lb3f

(Drawing: Hallway, by Marcie Christa Shawn)

Have you ever wondered how the freedom of choice that humans experience could be programmed into a robot?  In my sci-fi book This Moonless Sky, the species who call themselves the Communicators (a translation of their own word ‘Stcho’s!s!s!,,a’sh!sh!’) had to do this to outfit themselves for space travel.  Travelling through space in biological form is completely impractical – the only realistic kind of interstellar space travel is all-machine.  After much experimentation, they managed to program their entire personalities into machine form.

In this scene, though, we see one of them, Yith, now aged around 70 million Earth years, in a younger, human form.  He and the other members of his crew had transported the narrator, Marrik, in cryogenically frozen form from Earth to a new planet.  For a few months at the beginning and end of the journey, Marrik was in his normal human form, and he and Yith took a liking to each other.  While Marrik was frozen for the 400,000 Earth years needed to travel the 40 light-years between planets (you can calculate this on a napkin yourself), Yith completely cryo-mapped the neural patterns in his brain and also used some of his genetic material, with modifications, to make a human body for himself.  He could see from Marrik’s brain contents exactly what Marrik’s perfect partner would look like – a young man a year or two younger than Marrik’s 17 years.  After about 7 years of accelerated production of a suitable body (growing and maturing it at about twice the usual rate), he transplanted a copy of as much of his personality as possible into its neural nets and – after at first being rejected, basically for being too perfect – succeeded in becoming Marrik’s boyfriend.

In this scene they are in a region of the new planet that has a transplanted African culture (there called tsKorabaatse or Fuolbah, here on Earth known as Fulani), talking to the local traditional chief, the modibbei.  He’s wishing them off on the next leg of their journey, and by chance in their conversation, the topic of free will and robotics comes up.

(Glossary:  the slang English neutral ‘they’ indicating a single person, either male or female, is replaced by the local English ‘sey.’)

***  *** ***

“To wish you good luck in the Qodra,” the moddibei said, “my only bottle of sQodravtse sparkling white wine – champagne from the mysteries beyond our borders.  To the success of your trip, toasted in this elixir.”

We each got a proper flute of it.  Apparently there was quite a royal glassware collection tucked away someplace.  The gentle snap of the fizz threw faint alfalfa honey up at my lips even before they met this loving wallop of one flavour after another, green grape to sugar-apple to something that almost wanted to become a blackcurrant but hovered on the edge and changed back to gooseberry.

“I love watching your eyes when you’re having something you really like,” Yith said.  “It’s like a movie.  First sip, five scenes, different drama in each one, all done in one loooong take, and cut!  Set up for second sip!”

“This is gonna be one of my favourite films ever,” I said.  “You should watch some of yours.”

“I will, don’t worry.  But watching you is more fun.”

“Yes, but now I’m all self-conscious; the movie won’t be the same!  I’m in a Dauw-Lewrou room now.”

“Ah so,” the moddibei broke in, “you claim to be naïve (I had indeed told him I was still green in sDiyyanantse [local part of the planet] culture) and yet you speak with knowledge about my predecessor and his room.”

He had a point.  Even apart from schoolwork, I was soaking up the culture of my new land at a rapid pace.  I guess anyone would.  I probably don’t even sound like myself any more when I speak English.  That planet is gone, baby.  The Dauw-Lewrou Room is a simple philosophical thought experiment that was suggested by one of the first moddibeis of Korabaas, who took the made-up name Dauw Lewrou.  The name means ‘on the moon’ in tsKorabaatse.   He knew perfectly well that he wasn’t on the moon, so I think he intended the name as a pioneering post-ironic jest.

… The room’s main feature was that it had nothing in it whatsoever, just a hallway entering in the middle of one end, and an identical hallway leaving from the middle of the other end.  In the rest of the room, everything to the left of the entrance hall’s opening portal was identical to everything on the right.  There was nothing to distinguish the two sides of the room.  The experimenter would then instruct a number of people to run into the room, run over and touch the side wall of their choice, and run out the exit hall.  When they got out, they could try to answer a question they weren’t prepared for – why they had chosen one wall versus the other?  Underlying this procedure was the question of whether at least some of the people doing this would make the choice of left vs. right entirely by a simple exertion of free will.

Medieval Earth people had contemplated a somewhat similar problem called Buridan’s Ass, apparently a satire of the philosophy of Jean Buridan from the University of Paris in the 1300s AD.  The donkey in question was supposed to be equally starved and thirsty, and placed midway between a pile of hay and a bucket of water.  It would not be able to decide which of the two equal temptations to turn its head to first, so it would die, in place, of thirst and starvation.

Dauw Lewrou’s room was considerably more complicated than this.  He thought it likely that a person would always be able to choose between left and right on seir first pass through the room.  Some people might have a prior policy of preferring left or right in such cases.  Others might be influenced by other prior circumstances, like being left-handed and preferring to pivot on their right foot.  And some might be influenced by momentary circumstance.  As the Book of Power recapped it: “For example, sey may happen to arrive in the room on a particular foot, and may use that foot to pivot toward the wall opposite it.  Or sey may arrive in the room with one shoe slippery or differentially worn, or with a dust mote in one eye, and may base seir departure from the middle on one of these tiny physical vagaries.”  Still, Dauw-Lewrou proposed that at least some people, if you tested enough, would arrive in the room without significant influence from any of these factors, and would have to make an arbitrary choice.  They would discover that they have a faculty capable of doing that.  Free will, you could call it, or the faculty of arbitration.

On the person’s second pass through the room, though, questions of self-reinforcement would come in.  In that situation, the runner remembers seir previous choice and has to decide to go with it or against it.  Let’s say, in the first pass, sey chose to go left.  Sey has now probably lost seir freedom to just choose right vs. left and has to meta-decide whether to go with ‘left like last time’ or ‘right unlike last time.’  The decision may then jump up to the next meta-level:  ‘should I have a constant policy about this and if I do, should it be for variety, consistency, or a feeling of randomness?’  And yet another level may come in: ‘do I need to think about all this?’  In zen terms, the person has lost seir pure ‘beginner’s mind’ and has become all complicated about the whole thing.  This tends to slow sem down and make sem awkward, perhaps even slightly annoyed.  If sey gets around the problem, let’s say, by making a meta-decision always to go left, seir freedom in the room has been given up completely from that point onward.  Can sey ever get seir pure free choice back?  It seems very difficult to do.  It may still be operating on one of the meta-levels, but you can’t see it in the room.  If you do statistics on people who are trying to make random choices – let’s say in picking numbers between one and ten –  you find out that they cannot succeed.  Mathematical randomness eludes them; they always have a bias.  But could a master philosopher or a mystic achieve this?  No one knows.

So here I was with Yith watching me, and having to decide how to react to my second sip of sQodravtse champagne.  Should I react like I react, however that would be, or should I react for him in some way, like exaggerating the variety of reactions per sip?  Should I try not to show a reaction?  My spontaneity was lost – the champagne movie was over.

Meanwhile Yith was grinning away.

“Yes, he knows all about your predecessor, moddibei,” he said.  “And Marrik, I can have just as much fun watching your Dauw-Lewrou movie.  You think too much.”

“He’s right about that last bit,” I said to the moddibei.  “I think Dauw-Lewrou built that blasted room just for me to get lost in.  Now I have no idea what to do when I take my next sip.  Poker face? Clown face?  Humphrey Bogart sucking it in with man-of-the-world wisdom?”

I sipped anyways and made a wince of pain and pleasure.

“I mostly just tasted my own thoughts, but the wine underneath them was great.”

“We won’t watch any more,” the moddibei chuckled.  “You can vacillate in peace.”

Yith gave me a big kiss on the cheek right there.  Even Saraa [the modibbei’s wife] reached around from two seats away and ruffled my hair.

You just never know where your charisma is going to lie.

Then Yith partook of his champagne flute and I watched him.  He didn’t look at me and took the liquid with the same alfalfa-honey sweetness that it must have greeted him with.  Despite his complicated age situation, there was also a trace of mischief in his eye as you might see in a teenager from Earth who unexpectedly finds himself sipping high-end champagne like a junior Rothschild.

“You don’t look as complicated as I feel,” I said to him.  “How do you stay simple?  Young body?  Communicator reasonableness?”

“Oh well,” he said. “We had to decide way back when about decision-regress.  You know, when you’re programming a machine to address opportunities and weigh advantages, you always need to deal with that stuff.  You get in a Dauw-Lewrou situation and you not only have to decide – you have to choose a moment to decide.  You may delay, you may ponder.  So you have to decide to decide.  Or again, maybe you think you should decide quickly.  You may have to decide to decide a certain way, like quickly, or decide to decide that, what the heck, it’s just a room, it’s not so important.  If speed isn’t important, what then makes you decide to decide at a given moment?  Clearly you simply have to decide to decide to decide.  So there you are at level three of decision regress.  And there’s nothing to stop you from going further up the stack of more levels of deciding to decide to decide to decide.  Well, your human brain has a built-in safety buffer about this, so you can only go to about seven levels of decision-regress.  The process uses a part of your brain much like the buffer that temporarily remembers phone numbers.  Just a few humans who have different brains lose the barrier – calculating geniuses who can do any arithmetic instantly but who have a hard time with decisions, and get stuck in fixed policies to arbitrate their arbitrat [I think he was referring to some kinds of autism].  For example, they may always have to have exactly the same schedule, and sleep with their toy in exactly one spot on the bed.  That’s because otherwise, their decision-regress is limitless. They also can’t handle the other big kind of decision-regress, deciding to let their self-fulfilling prophecy be engaged by another person’s self-fulfilling prophecy.  Your people say these special ones have difficulty with human interactions and they can’t understand emotions.  Actually, they can – they experience all the emotions themselves.  But you have to decide that another person’s emotion is valid and identifiable with your own, and that you should commit to it as a reality.  It’s a matter of faith.  Faith is the ultimate Dauw-Lewrou room of decision regress.  How do you decide to have faith in someone or something and at just what moment do you decide it?  Well, you see, we had to program ourselves as opportunitarian machines, so we had to solve all these self-reinforcement problems.”

I looked at the fuzzy-lipped, velvet youth who was giving me this exposition and restrained any appearance of doing a double-take.   Did the verbiage match the downy cheeks, the lightly wafting sandy-blond eyelashes, the slaty-green-blue irises that contemplated me with such obvious affection?  I guess I must know that he has all this stuff inside him, I thought.  Somehow, it doesn’t quite make him a stranger.  Something in there really likes engaging my arbitrat, that’s for sure.  I wonder if I can acquire further insight into it by deep kissing.  Topic for later.

“So what did you do?” I asked.

“Our mechanical minds are very fast, but we still cut off our decision-regress at 677 levels.”

“So you decide to decide to decide to decide times 677?”

“Potentially. I mean, I’m human now, but I used to do that.”

“What made you guys decide on 677 levels?”

“It was an arbitrary decision made at the 676th level.”  Yith’s poker face wobbled into a smile.

Ach.  Communicator humour.  There’s really nothing like it.  Something somewhere in me had to giggle even though the joke was outrageously obscure.  At least, to a wad of meat like me.

“Can’t you just answer the question seriously?” I asked, with my smile broadening in spite of my objections.

Then he said something that I think may be quite profound.

“Marrik,” he said.  “There are some things that really can’t be taken seriously.”

*** **** ****

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s