Science fiction Reviews

Machines Like Me by Ian McEwan reviewed by Anil Menon: Back to the present


How does one write a sci-fi novel such that it won’t be mistaken for sci-fi?

Adam, the humanoid robot in Ian McEwan’s Machines Like Me, is fond of books in general and Philip Larkin’s poems in particular. So the chances are reasonably good he would’ve come across Larkin’s rejection of impersonating an imagined self: “I don’t want to go around pretending to be me.” This would leave Adam with a conundrum. Adam is designed to impersonate. The robot is most himself when he best pretends to be a human being. What is Adam to do? It’s a fascinating question. A science-fictional question.

Unfortunately, his creator Ian McEwan seems to have been motivated by another dilemma: how does one write a science-fiction novel such that it won’t be mistaken for science-fiction? A house divided against itself cannot stand. Neither can a novel.

Charles Friend, the novel’s 32-year-old narrator, lives in London. It isn’t our London. It isn’t our world. The year is 1982, Britain is about to lose the Falklands War, AI has made spectacular advances, Alan Turing is still alive, and Charles has just taken possession of Adam, a 180-pound robot, who looks like “a Turk or a Greek”, runs on a rechargeable battery, and costs £86,000. Charles is in love with Miranda, who is 10 years younger, lives in the flat upstairs, and is “a doctoral scholar of social history.” Which differs, I suppose, from majoring in “unsocial history”.

McEwan’s interest in plot, an atypical concern for a literary writer, though a welcome one, is evident in all his novels. But in this novel, there is none of the finesse we see in The Innocent or Sweet Tooth. Instead of keeping a razor-sharp focus on the relationship between Adam, Charles and Miranda, McEwan introduces a revengeful villain from Miranda’s past and an unwanted child. The story wastes considerable page and time in chasing the wrong conflicts.

Machines Like Me; Ian McEwan, Jonathan Cape, ₹699

Machines Like Me; Ian McEwan, Jonathan Cape, ₹699  

The problems aren’t all structural. For a novel deeply interested in personhood, the characters are little more than biodatas with legs. Arbitrariness is mistaken to be a mark of human spontaneity. Charles falls in love with a woman he hardly knows, buys an incredibly expensive robot, and offers to raise someone else’s child: all these decisions are made suddenly and arbitrarily.

When Adam tells Charles he loves Miranda too, Charles’ response is a series of facile details: “My pulse rate didn’t increase, but my heart felt uncomfortable in my chest, as though mishandled and left lying at a rough angle.” And Adam’s final scene, interspersed with forlorn aposiopesis, info dumps and Larkin-inspired haiku, is so protracted, so filmi, so ridiculously bathetic, every excess fold of mine jiggled with helpless laughter.

To my mind, McEwan’s style is a love-child of the styles of Robertson Davies and P.D. James. There’s this perennial struggle between the overt exposition of Davies and the calm observation of P.D. James. In McEwan’s novels like The Children’s Act, The Child in Time, and On Chesil Beach, the mother’s insightful style prevails. In novels like Solar and Saturday, it’s pater and his pointer who wins the day. In Machines Like Me, pater is completely out of control. There are pages and pages of exposition. Charles dumps the same biographical info in Chapter One and in Chapter Seven.

Similarly, he meets Turing twice, and in both meetings, Turing undertakes to deliver a compressed history of How AI Succeeded Thanks To Me. Poor Charles even reacts the same way: he wants to let Turing know he isn’t ignorant, but doesn’t. It’s as if McEwan wanted to use cut-and-paste but accidentally used copy-paste instead.

Defective psychology

It doesn’t help that McEwan’s alternate world is an implausible mess. Partly this is because McEwan gives many details, unwisely and often carelessly, to establish plausibility. For example, we learn that Charles’ car, a mid-60s’ vehicle, is a “British Leyland Urbala, the first model to do 1,000 miles on a single charge.” Since the novel’s 1940s more or less mirrors our 1940s, it means that in 20 years, McEwan’s world has gone from cars that run 15-20 miles/ gallon to a 1960s’ electric car with almost twice the mileage of a 2018 Tesla! On the other hand, we’re also told that Adam takes 16 hours to recharge on a standard 13 amp socket. Since Charles has already told us that “At thirty-two, I was completely broke,” it can only mean electricity costs next to nothing.

The real technical problem is that McEwan does not want to use a near-future setting. So he sets his novel in an alternate 80s’ world similar to our world, except for an accelerated development of technology after World War II. Turing is trivialised into a totalising genius responsible for practically every advance in computer science. The robot’s existential dilemmas were new in the 1940s, when Asimov wrote his tales, but McEwan seems to think they’re brand new. Turing sombrely informs Charles that “we don’t yet know how to teach machines to lie,” but didn’t robots supposedly pass the Turing test in the 1960s?

My criticism is not about “getting the science right”. It is about getting the psychology right. The Great Chain of Servitude — slave, serf, servant, employee and devotee — now includes a new subaltern: the robot. The stories we tell about robots are stories about our evolving understanding of personhood and servitude. History matters. By disrespecting history, McEwan reduces this understanding to a caricature. The result is a tedious novel, incompetently written and incoherently conceived.

The writer’s novel, Half of What I Say, was shortlisted for the 2016 Hindu Prize.

Why you should pay for quality journalism - Click to know more

Recommended for you
This article is closed for comments.
Please Email the Editor

Printable version | Dec 6, 2019 1:06:19 AM |

Next Story