Short Story: “Self Portrait” by Bernard Wolfe

(Cover artist uncredited. Galaxy, November 1951.)

Who Goes There?

For someone who wrote so little SF, Bernard Wolfe (no relation to Gene Wolfe) is very much someone worth rediscovering, if for no other reason than that he brought a worldview to the table that was quite rare in his lifetime, and which has since become only somewhat less so, even as the field has become more progressive in its collective politics. Wolfe trained as a psychologist, and seemed to be one of the few proper Freudians of his time to have written SF; he had even himself undergone psychoanalysis in 1950, about the same time he would’ve written today’s story. He was also a leftist, specifically a Trotskyist, to the point where as a young man he even acted as a bodyguard for Leon Trotsky for several months, although he had left his post over a year prior to the latter’s assassination. While Wolfe would come to no longer identify as a Trotskyist, no doubt at least in part due to the red scare in the years following World War II, he seemed to remain committed to the causes he thought were worth fighting for. On top of being a psychologist he also made some extra money teaching at university; and while he only wrote one SF novel and a handful of short stories, he did write a decent amount of work that wasn’t SF. ISFDB can be frustrating with not cataloging an author’s non-genre work, since here it gives the impression that Wolfe had vanished from the face of the earth between 1952 and 1960. Most notably he wrote a fictionalized account of Trotsky’s last days titled The Day the Prince Died, or The Death of Trotsky.

Wolfe made his SF debut with “Self Portrait,” whose themes he then expanded upon for his sole SF novel, Limbo, published the following year. “Self Portrait” is densely packed, maybe too much for its own good, but it’s fascinating, there’s a good deal to unpack, and it’s a good example of the kind of socially aware SF that H. L. Gold wanted for Galaxy, his newfangled magazine that was already forever changing the way we understand the genre. You can see why Harlan Ellison, when commissioning stories for Again, Dangerous Visions, was eager to nab a Wolfe story (he ended up getting two of them), despite the latter’s inactivity.

Placing Coordinates

First published in the November 1951 issue of Galaxy Science Fiction. It’s been reprinted in The Robot and the Man (ed. Martin Greenberg, not to be confused with Martin H. Greenberg), Second Galaxy Reader of Science Fiction (ed. H. L. Gold), Tomorrow and Tomorrow (ed. Damon Knight), and We, Robots (ed. Simon Ings). It seems to have fallen out of copyright in the US, so you can read it on Project Gutenberg.

Enhancing Image

Oliver Parks has recently been assigned to Princeton University, to work in its cybernetics department. The story is set over the course of a few months, in the fall of 1959, which was then the near-future. Parks has taken on the hefty task of designing a life-like robotic leg, after his boss was impressed with his “photo-electric-cell” insects, often referred to as bedbugs—little robotic bugs of Parks’s design that look and more importantly act like real bugs, not because of the exterior but what goes on inside. The nerves. Prosthetic limbs are by no means new; in fact, if you think about it, basic prosthetic limbs have been around since at least the golden age of piracy. Wooden pegs for legs and all that. But then there’s the issue of developing a limb that can look and act like its real human counterpart. Perhaps even better than the real thing. Thus we have Parks’s subject, named Kujack, a Korean War veteran who had both of his legs blown off. (It’s worth noting that Wolfe probably wrote “Self Portrait” in the fall of 1950, so that the Korean War had only been going on for maybe a handful of months, making it surely one of the first SF stories to mention that conflict.) There’s also Parks’s assistant, Goldweiser, who despite his position is openly an egomaniac, and I do have to wonder if it’s a coincidence that Goldweiser reminds me of the similarly named H. L. Gold. Gold, for better or worse, worked closely with his writers, and he quickly gained a reputation for being a control freak. I have little doubt that Wolfe would’ve found Gold a pain in the ass to work with.

And then there’s Len Ellsom. Parks and Ellsom go back quite a bit, not least because there’s the implication Ellsom had “stolen” Marilyn, who was Parks’s fiancée. Ellsom also has a habit of calling Parks “Ollie,” which irritates him, although it’s unclear if Ellsom is doing this innocently or not. Ellsom is not a cyberneticist; instead he’s working on a supercomputer that can beat even the best human player in a game of chess—a computer so smart and tactically oriented that it could win a war all by itself. This was written in the days long before the microprocessor, and before the real-world evolution of the computer that was designed to play chess against professional opponents. You might have a fuzzy memory of the supercomputer Deep Blue playing against world champion Garry Kasparov. In case it wasn’t obvious, we’re talking about machine learning, which nowadays tends to be just called AI, although calling it AI isn’t really being accurate. Deep Blue was not able to write poetry or commit suicide (after all, being able to write poetry and commit suicide are the two biggest signs of human-like intelligence), but it was able (indeed designed) to think strategically. There was a problem that needed solving (a game of chess) and Deep Blue could solve it. Parks considers machine learning to be easier than constructing an artificial limb because while Parks has to replicate the human leg, Ellsom is “merely” working with learning machines, which can replicate a certain function of the human brain, but not all of it.

That’s why the robot-brain boys can get such quick and spectacular results, have their pictures in the papers all the time, and become the real glamor boys of the profession. They’re not asked to duplicate the human brain in its entirety—all they have to do is isolate and imitate one particular function of the brain, whether it’s a simple operation in mathematics or a certain type of elementary logic.

There’s some evident jealous in Parks’s tone: it might be the thing Marilyn, but it could also be the fact that Ellsom has the more glamorous job. Things are not as straightforward as they seem, though, not helped by Parks being a bit of an unreliable narrator. “Self Portrait” is told as a series of diary entries from Parks, so that we’re hit with a double whammy of a first-person narrator who’s also bitching about his colleagues in real time. It also goes to explain Wolfe’s rather informal style here, since presumably Parks is writing these entries without the expectation of them getting published, and thus we see a side to him other people would not.

One night Parks and Ellsom have a bit of a night on the town, and Ellsom gets quite drunk; but then of course, he confesses he’s been drinking more than he should. The reason? “I saw a machine beat a man at a game of chess.” Three years ago. Machine learning is apparently a bit farther along in development than Parks had thought, and despite being involved in the development himself Ellsom is rather disturbed by it. He has a friend, Steve Lundy, who lives in Greenwich Village. “He’s a bum, you see, but he’s got a damned good mind and he’s done a lot of reading.” Lundy is pessimistic about the development of a supercomputer that can beat a man in chess, and it becomes hard to not see why. Incidentally Lundy is also a leftist, having fought in the Lincoln Brigade during the Spanish Civil War. It seems to me that of the characters featured in “Self Portrait” that Lundy might resemble Wolfe the most, the irony being that we don’t actually see him until the end. Speaking of Greenwich Village, Ellsom brings up Marilyn again, apparently guilt-ridden, and explains to Parks, in so many words, that he shouldn’t feel too bad about having lost Marilyn, on account of her being a huge slut who had fucked every man in the Village who would have her. In the Village she tried being an artist, like every other bohemian, but apparently gave that up to go back to doing a tech job at IBM. Despite her sluttery, Marilyn is said to be a very intelligent woman, and there’s the hint that Parks still sometimes thinks about her. “You know,” he says, “she helped lay out the circuits for the first robot bedbug I ever built.”

If it seems like a lot is going on without my ever feeling the need to go through the plot beat by beat, that’s because it is. “Self Portrait” is about 9,300 words according to the Gutenberg text, making it a solid novelette, but even so it feels squeezed in, almost like the abridged version of a much longer and deeper narrative. I would have preferred if Wolfe had doubled the length and turned it into a novella, which would’ve given his ideas more room to breathe. If contemporary reviews are anything to go by his novel Limbo, which expands on the ideas presented in “Self Portrait” but not the plot, suffers from the opposite problem: at some 430 pages it’s freakishly long for a ’50s SF novel. Wolfe is a lot more cerebral (not to mention left-leaning) than some other ideas-heavy SF writers of the period like A. E. van Vogt, but he still has the problem of not being a very good storyteller. It strikes me as apparent that “Self Portrait” probably should’ve been narrated by Ellsom, not Parks, although it’s possible that by sticking us in his shoes Wolfe wanted to make a point about the kind of person Parks, which is to say a square. A conformist. Ellsom calls him a Boy Scout, which Parks takes as an insult. Something you realize about Parks is that he’s kind of an asshole, and also prone to lying for the sake of keeping up appearances. Not to get ahead of the plot a bit, but the ending solidifies Parks’s status as anti-hero, in that he is anti-heroic. The only reason he doesn’t come off worse is because Goldweiser is more openly unlikable; and anyway, while Parks is not exactly a good person, Wolfe posits that the likely future, in which the US builds a computer that can conduct a whole war, is leagues worse than one petty cyberneticist. It’s bleak, if you think about it.

There Be Spoilers Here

Parks finds out that Ellsom’s recklessness has gotten to the point where he’s not only hanging out with Kujack, but sharing government secrets with him. This is a development that mostly happens offscreen, which can make the reveal of Ellsom and Kujack both technically traitors a surprise in the sense that the reader simply would not have expected such a development to occur. It feels random. There is, of course, a bit of foreshadowing, about how something is off about Kujack, and there’ve been mentions of Ellsom’s alcoholism; but then both counts can be dismissed, or at least downplayed, on account of us only being told about this through Parks’s rather biased perspective. The implication of the twist and the two men’s subsequent arrests (although in a humorous sequence of events Kujack isn’t arrested until after the team gets to show to the public that Parks’s artificial leg is a success, for the sake of PR) is that Ellsom may have turned to hard drinking out of a crisis of conscience, given that he’s been helping with a project that could potentially lead the US and Soviet Union ending the world. (Less of a Dr. Strangelove and more of a Fail Safe situation, by the way.) Ellsom had also been yapping about the project with Lundy, who in turn has been running an anarchist magazine named after the supercomputer Emsiac. Things end badly for about half the cast (there are too many characters here, really), although not Parks, who comes out unscathed and is even able to reunite with Marilyn at the end. The conformist lives happily ever after; meanwhile the characters who are framed as more sympathetic, indeed those whom Wolfe likes more, are punished.

A Step Farther Out

I would be curious to read Limbo, since it seems like the kind of overly ambitious mess that was sadly rare in ’50s SF. “Self Portrait” is the only Wolfe story to appear in a genre magazine; all the others would be published either in mainstream outlets or anthologies. This story doesn’t count as cyberpunk, mind you, but it does feel like a distant precursor to the movement, by a good thirty years, in that it deals with cybernetics (well, duh), but also its thematic concerns anticipate the anti-capitalist and anti-war stances of most classic cyberpunk writing. In that sense Wolfe was ahead of his time, although he wrote so little SF and from what I can tell he could never quite tap into his own talent to produce a masterpiece.

See you next time.


4 responses to “Short Story: “Self Portrait” by Bernard Wolfe”

  1. I was in the mood for a great old science fiction short story and “Self Portrait” fit the bill. It was very satisfying. Excellent review. I wish I could describe stories like you, especially seeing into character motivations.

    “Self Portrait” was incredible well-written for SF at that time. Plus, I was surprised by how much Bernard Wolfe knew and guessed about AI in 1951. By the way, I was born November 25, 1951, and this might be the best story I’ve read from a SF magazine that came out with my birth month and year.

    Liked by 1 person

  2. I also meant to say that back in the 1960s when I first haunted used bookstores I’d always find copies of LIMBO by Wolfe, ALAS, BABYLON by Pat Frank, and ON THE BEACH by Nevil Shute. Evidently, they had been bestsellers in the 1950s and there were tons of old copies.

    Never read LIMBO, but I eventually read the other two. Last year I bought Kindle edition of LIMBO but still haven’t read it.

    By the way, I think I’ve read “Self Portrait” before but just can’t remember where. I own that issue of GALAXY. I also have it in two anthologies.

    Like

  3. [1] I’ve read Wolfe’s LIMBO 90 (as it was titled in the UK paperback edition in the 1960s) and have favorable memories.

    Granted, this was decades ago and, on the one hand, I was a fourteen-year-old who was relatively undiscriminating in my search for SF. On the other hand, though, I found Wolfe’s novel straightforward in a mode adjacent to the kind of social criticism that Pohl-Kornbluth (or for that matter Bester’s STARS MY DESTINATION and much 1950s American SF) did. I didn’t bounce off in the way I did Vonnegut’s CAT’S CRADLE, for instance.

    So LIMBO is probably quite readable and might be worth your while.

    [2] You write: “instead he’s working on a supercomputer … so smart and tactically oriented it could win a war all by itself. This was written in the days long before the microprocessor, and before the real-world evolution of the computer designed to play chess against professional opponents.”

    These ideas were around from the very start. Take a look at this TIME magazine cover from 1954 with a computer general/admiral, for instance —

    https://www.pinterest.com/pin/1954-pentagon-computer-artzybasheff–655555289506296924/

    In 1946, in fact, the Pentagon sponsored the building of the first electronic stored program computer, John von Neumann’s IAS machine at Princeton, specifically to run the hydrodynamics calculations for the Ulam-Teller ‘Super,’ a three-stage device in which a radiation-imploded fusion second stage would boost an initial fission explosion, producing in the third stage a thermonuclear blast hundreds of times greater than the atomic bomb’s.

    Likewise, cybernetics’ founder, Norbert Weiner — who was very clear about the future relationship between machine learning and whatever biological nervous systems did — only got wide-scale acceptance for cybernetics with an aiming device for anti-aircraft guns in WWII.

    Similar theories to Weiner’s about learning and neural networks, both biological and mechanical ones, were prevalent at least since 1943 and the seminal Pitts-McCulloch paper A Logical Calculus of Ideas Immanent in Nervous Activity. (Pitts is a fascinating story, a homeless genius who corresponded with Bertrand Russell.)

    So the Pentagon was interested in cybernetizing war from the beginning. As regards the first microprocessors not having arrived in 1950-51 when Wolfe wrote this story, remember that the transistor had been invented in 1947 and a microprocessor is nothing but a bunch of transistors printed on one chip. There’d been a general recognition that the vacuum tubes in the first electronic computers like von Neumann’s wouldn’t permit advance and transistors were being proposed as an alternative, if only an architecture could be figured out to put a great many of them together. Viola; the microprocessor chip.

    Wolfe was simply right on top of where the smart scientific and engineering thinking was in 1950. Probably more so than anyone in John Campbell’s Astounding, actually, except Wolfe would never have published there because Campbell was a fascist by then, as Donald Wollheim said at the time, even if he’d not been one before.

    Thus Galaxy provided an outlet for this new kind of science fiction in the 1950s, as you say, and enabled what most of us probably feel — as Barry Malzberg did — was the real Golden Age of American SF

    Liked by 1 person

Leave a comment