Tuesday, 28 June 2011

Do Androids Dream of Electric Sheep?

Writing about what it means to be human made me think of Philip K Dick's lovely science fiction novel, Do Androids Dream of Electric Sheep?.  The title alone has got to be worth the price of the book. It poses a tricky, if hypothetical, problem which is not that different from the problem of post-humanism.

The story centres on two humans.  Rick Deckard is a bounty hunter whose job is to destroy escaped androids.  The intellectually disabled JR Isidore is a delivery boy for a company that repairs electronic animals.  They live on an Earth that is a virtual wasteland, where almost nothing survives except humans and even these in rapidly decreasing numbers through mass emigration to the outer planets. 

In this lifeless world, every human dreams of owning a real animal, but these are such rare and expensive items that most have to settle for incredibly lifelike electronic substitutes.  These dreams provide a deep emotional core to the novel.  Deckard, already the owner of a electric sheep, takes his flying car out into the wilderness for a little time alone and finds what appears to be a real, living toad.  He carefully catches it and takes it home to his wife, only to be shattered as she ruefully flips it over and opens its battery cavity.  JR has better luck, finding a real live spider in the abandoned unit block where he lives and taking it back to his own unit.

In this world there are also human-shaped androids, so lifelike as to be almost indistinguishable from real humans.  They look exactly like humans, they sound like them, they smell like them, they are as intelligent as them and they even behave like them.  Used widely as servants on the outer planets, these androids are outlawed on Earth.  Nonetheless they frequently escape and try to blend in with the Earth population, and it is Deckard's job to find them and destroy them.

This is not easy, as androids have no trouble blending into the human population.  Short of pulling them apart, the only detectable difference is that androids have no empathy.  They know what it is, and can mimic it, but they do not feel it.  When Rick identifies a suspected android he has to administer an "empathy test" in which he hooks it up to a monitor and meaures its response to various emotional stimuli.  A human will have an instant physiological reaction to shocking ideas and images.  In an android this reaction will be ever so slightly delayed, as they first process the idea, and then consciously create the appropriate response.

This difference is crucial, as we see when a group of escaped androids move into JR's apartment block.  JR takes them for humans and befriends them.  At first things go fine, but then they find his precious spider and slowly torture it to death in front of him, oblivious to his immense distress.  JR himself is rescued when Rick tracks down the androids, but his relief is tinged with sadness for the loss of his "friends".

This is, of course, a version of the Frankenstein story, and it poses a similar moral question for us, should we succeed in creating artificial intelligences that really live.  In Dick's view, our creations may be monsters, highly dangerous in much the same way as Frankenstein's monster.  Yet the book's title insistently asks us to empathise with them.  Perhaps they do dream of electric sheep.  They certainly dream of freedom, or else why would they escape so regularly?  Like Frankenstein's creation, they dream of being truly human.  Could they be, or will they be always wholly other?

No comments: