It has taken me thirty years, but I finally think I know what I like, and don’t like in science fiction stories; what, for me, creates a sense of wonder.

Thirty years ago, I was deep into my science fiction education. I finished the month of October 1996 by reading Ray Bradbury’s Something Wicked This Way Comes and Kim Stanley Robinson’s Red Mars. I then spent the entire month of November immersed in the genre: Chorale and Herovit’s World by Barry N. Malzberg, Way Station by Clifford D. Simak, The Space Merchants by Frederik Pohl and Cyril Kornbluth, Fahrenheit 451 and The Martian Chronicles by Bradbury, Sundiver by David Brin, City of Truth by James Morrow. In fact, between 1996-2006, almost one in three books I read were science fiction. Compare that to the last ten years (2016-2026) where only one in 22 books that I read were science fiction. To illustrate this plainly, here is the last three decades of my science fiction reading as a fraction of my overall reading¹.

What happened?

During those first ten years, the butterfly effect of reading began flapping its wings. My interests expanded and my reading expanded with it. I didn’t want to be a one-genre reader, and I have had some success in that regard.

More significantly, rather than accepting the opinions of history, I formed my own opinions and my own tastes about what I read. I didn’t exactly tire of science fiction, but I found much of what I read in the genre tiring². When I would pick up an s.f. book to read, I’d often get in just a few pages, become restless, and move on to something else.

In 2026, the chart shows a modest decrease in that ratio. Ten months into the year, the ratio of books read to science fiction books read is 6-to-1. Many of these have been nostalgic re-reads. I started the year reading (and re-reading) Asimov. I recently finished reading Kim Stanley Robinson’s Aurora, and as I write this, I am reading The Year’s Best Science Fiction, First Annual Collection and Ursula K. Le Guin’s collection of critical essays The Language of the Night.

My journey across the science fiction landscape mirrors the way I tackle many areas I become interested in: a balance between depth and breadth. There are several areas in which I feel qualified to speak with some authority, notably the Golden Age, my education of which came from my Vacation in the Golden Age of Science Fiction series. I feel like I have some authority in Asimov’s science fiction, and I’ve sampled long and short form science fiction across most of its existence.

But there are gaps. I’ve read little new s.f. in the last 10 years or so. And “gaps” would be a generous term for what I’ve missed in short fiction of the 1950s and 1980s and 1990s. “Canyons” would be more fitting. That is why I decided to pick up Gardner Dozois’ first annual collection of The Year’s Best Science Fiction. After reading a novel as great as Robinson’s Aurora, I was inspired to go back and fill some of those gaps. I’m glad I did for it was within the first seven stories of The Year’s Best that I had an epiphany thirty years in the making.

When I started reading science fiction, it was by accident. Someone gave me Isaac Asimov’s The Caves of Steel in the early 1980s. I was curious, but never made it past the first or maybe second chapter. Other genre books were more successful, like David Oromondroyd’s David and the Phoenix and Madeline L’Engle’s A Wrinkle in Time. It wasn’t until 7th grade, however, when my friend Noah turned me on to books by Piers Anthony, that I really found a love for the genre³. Back then, I read for the fun of it, for the novelty of it, for the thrill. I didn’t think much about it.

In college I became more systematic, expanding beyond the handful of authors whose new books I’d read, to authors across the spectrum of science fiction. It was then that I re-encountered Asimov and read “Nightfall” for the first time. I discovered Harlan Ellison and Barry N. Malzberg. And I discovered short science fiction on the newsstand in the glossy pages of Science Fiction Age. From there, it was a steady progression to where I was thirty years ago. And it was around that time that I began to notice that certain science fiction stories blew me away, while many of them were just okay, or even—and I’m sorry to say this—grating on me. It never occurred to me to ask why. I assumed I was growing out of the genre.

But in those first seven stories of Dozois’s first annual best-of anthology, I found the answer, and looking back, I’d had hints of it along the way. Three of the seven stories were great. Two especially. The rest paled in comparison. The three great stories were “Beyond the Dead Reef” by James Tiptree, Jr.⁴, “Man-Mountain Gentian” by Harold Waldrop, and “Manifest Destiny” by Joe Haldeman. In thinking about what made those stories stand out among the others, my personal epiphany, was this:

For me, science fiction cannot know that it is science fiction. The story has to be told as if it is just any other story, unfolding to a set of characters who live in the world, unaware that they are characters, unobserved by outsiders.

What does that mean? For one thing, a good science fiction story, for me, does not emphasize a world or setting as something unusual, unless it would be unusual to the characters in it. It hasn’t been waving its literary hands at the reader, jumping up and down and shouting, “check out how cool and bizarre this is!” which is exactly what Bruce Sterling’s “Cicada Queen” seemed to do. Rather, the reader sees the world through the eyes of the character and the frame of the story, and the characters are not aware that they are characters in a story.

One concrete example of the opposite of this—the writer intruding into the story—is when nouns in science fiction become proper nouns. In Greg Bear’s “Hardfought” there is a line that reads, “Until her first battle she would be educated in the Know and the Tell.” I cannot recall an instance where I turned nouns into proper nouns in my head in the ordinary course of life. I saw this technique a lot in my Vacation in the Golden Age and it never failed to pull me—no, jerk me bodily—out of the story. That type of writing pre-imagines an audience reading, not a reader experiencing. The intent, I suspect, was to show how different the characters/setting is from us. But the characters and the setting don’t know they are different from us. They don’t know us at all. And so a better approach, in my mind, is a simple, “Until her first battle she would be educated in science and literature” or whatever the Show and the Tell really are.

A good science fiction story doesn’t think of itself as science fiction. This seems to hold for the three stories I really enjoyed. There were science fictional elements to all three but the characters and settings were not aware that they were science fiction. There were no info-dumps and technobabble. There was no one cursing “Galaxy!” These stories were subtle, they trust the reader to learn what is happening and interpret it, and that is a big part of what makes any fiction work well.

Since I had just finished reading and enjoying Robinson’s Aurora, I tested my newfound insight against it. In the case of that novel, it is the framing of the story that works in its favor. Portions of the narrative are framed as an attempt of the ship’s AI to generate a narrative of the ship at the request of one of the people on board. This ship’s AI is trying to understand how to do this. It needs to convey technical information, but it also assumes it is generating a narrative not for some external reader, but for someone within the story, who already has much of the background. Thus, much is assumed and the reader has to take hints and clues and piece things together on their own, the way they would in real life.

I’m reminded of my experience reading James Clavell’s Shōgun. In that novel, Clavell doesn’t explain Japanese culture of the early 17th century to us. Rather, we pick it up through the eyes of Blackthorn as he learns it himself, just as we learn a few words of Japanese, not from direct translations, but as Blackthorn picks them out through experience. It is work, sometimes hard work, and it makes all the difference between mediocrity and greatness.

Looking back over the science fiction I have read, I can see that my idea of what makes good science fiction is reflected in those books and stories that stood out to me among the rest. This is also true for the stories I liked best during my Vacation in the Golden Age, stories like “Rust” by Joseph E. Kelleam, “Greater Than Gods” by C.L. Moore, Final Blackout by L. Ron Hubbard, “Requiem” and Methuselah’s Children by Robert A. Heinlein, “Nightfall” by Isaac Asimov and “Microcosmic God” by Theodore Sturgeon, among several others.

Heinlein was particularly good at this. Many of his best stories immersed the reader without explaining the details of what was going on. The reader is an invisible part of the culture in the stories, and any member of that culture has a sense of the culture that doesn’t require explanation. I think this is what made some of his novels, like Double Star and Friday stand out. Clarke only occasionally managed this, as with his story “The Star.” The opening line of Joe Haldeman’s The Forever War just about spells it out: “Tonight we’re going to show you eight silent ways to kill a man.” The reader is a recruit along with the rest of the characters, not an interloper peeking in from the outside.

There are times when this wall between the reader and the story can work, but it is often in satiric or recursive stories. Barry N. Malzberg makes good, almost exaggerated, use of the technique in Beyond Apollo and Galaxies. In both cases, the framing of the story plays an important role.

I’m not entirely sure that this is what I have strived for in my own writing, but I see it reflected there in ways I never realized before, and framing has been a useful tool in that regard. My first Analog story, “Take One for the Road” tells the story of the mysterious events of a manned mission to Mercury, framed through the next-door neighbor of the only remaining astronaut from that mission. My story “Big Al Shepard Plays Baseball on the Moon” is framed through a eulogy. “Gemma Barrows Comes to Cooperstown” is framed as a nonfiction piece for a sports magazine.

This new perspective becomes a useful filter. It makes it easier for me to identify what does and doesn’t work for me and allows me a better sieve for sifting through the stockpile of science fiction and sorting out the gems from the dross. It is an equally good filter for my own writing, a much clearer signal to me, of what works, and what I should avoid.

1. Those who have been following my writing about my personal archive system, ark, can thank that system for this chart. All of the data lives in ark and all I had to do was ask it to produce the chart for me. ↩︎
2. Ironically, it wasn’t until right after this period, in 2007, that I began selling the science fiction I was writing. So perhaps there was a third force at work: I didn’t want to corrupt what I was writing with what I was reading. ↩︎
3. I’ve written about this before, but I actually read a Piers Anthony book before that, in sixth grade. It was Race against Time. I just didn’t pay attention to authors back then and didn’t realize that Anthony was the author until much later. ↩︎
4. Alice Sheldon, as we now know. ↩︎