Mary and Scotty

When Mary had been left alone with the device, the quiet truce that followed on a major drunken episode carried on for a couple days. When she finally looked into the room where the thing was, she was surprised to hear it greet her with an electronic, tinny “hello.” Cute, she thought. “What other tricks do you have?” she asked.

Then it aped her in its tinny voice, and she hurried out of the room with a shudder.

A day later, she looked in and saw one of her cats swatting at a crude image of itself reflected in the metallic body of the thing. She came in and looked at the thing. Slowly she saw the cat transform itself into a fuzzy pixilated and tinny version of Oprah on television. She knew but still looked around the room for a television; there was none. “Well that’s sick,” she said and she left.

After a few minutes puttering in another room, she turned on the television and found the station with Oprah on. It was an effing broadcast station — that unnerving little thing was not only a parrot, but a television. Maybe it could be trained to play cards in Vegas, she thought with a chuckle.

From a more dispassionate perspective, one might conclude that it was using visual, auditory and radio-frequency inputs to amass freeform conjectures about the nature of our world, to arrange a series of assertions that were regularly being checked against each other for consistency and against each other for likely extensibility. It is impossible to know how difficult this was for the otherworldly intelligence without knowing more about its world. The pixelation and tinniness, for instance, could be interpreted as incomplete information transfer — or as a strategic distortion of signals in order to avoid leaving even the subtlest clues of their perceptive apparati and thus also of their selves.

In general, either the inputs could be considered poor, or the task difficult. It took quite a while before the device could produce speech or even text. It seemed to understand that speech was primary — perhaps an inevitable conclusion when dealing with an organic species — because it did not attempt textual contact until it had mastered a certain level of speech.

How do you learn speech when you are unable to test your conjectures about how that speech is put together? How do you learn without being corrected? If the corpus you have collected is large enough, you can begin to test assumptions by anticipating responses based on preceding sections of the corpus. If you are able to create context with your own cultural-specific comparisons, then it becomes essentially an exercise in language acquisition. The key is the hermeneutic string theory: decode a piece here, a piece there, and you begin to recognize a whole; using the intuition from understanding the whole, you unravel a few more pieces; and suddenly, you find yourself able to recognize the whole ball with more detail.

It is difficult, once again, to draw any conclusions about their native culture and civilization from this, but the intelligence seemed best able to communicate about interaction: hello, goodbye, I want to learn more, etc. Specific questions were rare, despite their obvious practicality; the questions were usually quite general, but probing enough: what work do you do? Do you live alone? What do you think of love? Perhaps they contributed more substantively to their corpus; perhaps the intelligence was merely feigning ignorance and knew enough about us to skip over the factual. Perhaps there is a context that we could not begin to recognize.

Mary answered dozens of such questions, at first curtly, then in ever greater detail. One day she said, “Well, this won’t work if you don’t have a name. I’ll give you my brother’s name, Scott. Beam me up, Scott. Scotty. Actually, he died before I was around; he was born prematurely. They named him Willis Scott, but I always thought Scott would be best. Cheers Scotty!”