The Best Science Fiction Books (According to Reddit)

MINOR SPOILERS for Accelerando

I thought Accelerando was great up to the part where it stopped making any sense (about the time they were fighting some alien invaders parasites that teleported from somewhere using communication channels). Maybe I'm just dumb, but I really couldn't keep up with the plot and it got very nonsensical so I stopped reading. Dealing with singularities, the divide between the physical and the abstract is always very fluid. While Stross dealt OK with the abstract reality, I thought his physical reality didn't make any sense to me. Hilarious extrapolations, and such.

This happens with more writers (and other artists) than many people think. I read Carlos Ruiz Zafón's The Shadow of the Wind and The Angel's Game and loved both: they're incredibly rich and detailed yet driven by plot; they have the action that a lot of unsophisticated readers crave and content that can still satisfy someone who's read 10,000 novels. Oh, and it made me laugh out loud.

U.S. publishers are now releasing his young adult novels. Two so far, two more to come. And they're terrible. Almost unreadable bad. No plot, cardboard characters, wild improbability poorly embedded in a paranormal universe. You can see flashes of his later skill, but his early stuff is lousy.

I said this in another comment, but I'll say it again here: I just started reading David W. Galenson's Old Masters and Young Geniuses: The Two Life Cycles of Artistic Creativity, (http://www.amazon.com/gp/product/0691133808/sr=8-1/qid=13150...), which is where Malcolm Gladwell stole / sourced his New Yorker article "Late Bloomers - Why do we equate genius with precocity?" (http://www.newyorker.com/reporting/2008/10/20/081020fa_fact_...). Galenson says that there are two basic modes for artists to follow: the experimentalists who try many things over long periods of time and eventually reach fullness, and the "conceptual innovators" whose sudden insight into a field fuels their work, which is often done at a young age. Under your reading, Stephenson would be the "experimentalist."

Diamond Age painted a very deep and rich tapestry of future vision.

http://www.american-buddha.com/scandark.authornote.htm

Also, Pynchon's prose is very rich - I expect popular lists to be dominated by lighter fare.

I particularly like The Star Diaries, The Cyberiad and His Master's Voice.

Likewise - This is how I recently found 'The Windup Girl' by Paolo Bacigalupi; thoroughly enjoyed this book and would highly recommend it.

Perfect for finding new interesting writers (Bacigalupi was one of the nicest discoveries)

[1]: http://en.wikipedia.org/wiki/The_Years_Best_Science_Fiction

... all of it beautifully unencrypted and unencumbered. What’s the catch? This disk and its contents may be copied and shared, but NOT sold. All commercial rights are reserved. That’s it.

I think the original three are so good because they are much less tainted by actual technology in the real world it seems, and it's sort of funny how Asimov, in some sense, describes Wikipedia and the utter importance of trade and consumption in ruling those around you, weaving in technology and concepts that even today sound like real science fiction; far fetched but attainable in the future. More importantly, Asimov does a good job in connecting you with the characters.

But ugh ugh ugh. Every single book thereafter gets worse and worse, and by the end it's so bad that I actually welcomed the blatant fanservice reappearance of you-know-who.

I'll toss in my recent favorite space opera: the Polity series by Neal Asher. Similar to Banks in that it has governing AIs, but less omniscient; good adventure stories, with a dash of horror in original hostile species, sentient and otherwise.

I was particularly struck by the concepts that came from the Atkins character, as the sole entity in the universe's military. The magnificent display of unbelievable power, not as planet cracking explosives, but weapons so precise that individual neurons could be severed from orbit, altering the course of entire conflicts. And the idea of the weaponry and strategy, though controlled by godlike machine minds, still needed the ok of a person to do the precision strike.

The only real weakness I felt was the Daphne character, she was flat, uninteresting and almost misogynistic in her treatment as an unbelievable ignorant and foolish female character. I found myself aggravated during her sections of the work, this stupid character in the midst of such magnificently crafted, almost baroque, prose and ideas.

Our (American) culture has changed so much that modern readers think "what's the big deal with this".

The drugs/orgies etc in Stranger in a Strange Land have much less of an impact on a generation that grew up with easy access to hard core drugs and hard core porn.

The book's most glaring weakness, however, is that many of its characters suffer from I'm-going-to-spell-out-precisely-what-I'm-about-to-do syndrome (Harkonnen, in particular, has a James Bond Villain-esque penchant for such speeches.)

Regardless, the book has been absolutely foundational to the genre. Nearly every convention of the modern space opera can be traced back to Dune.

It's a pity that no one's ever been able to crack the book's code on film. Understandable, though, as so much of the action is psychological and the drama internalized.

Personally, I found the first and fourth Hyperion books to be my favorites, and the middle two less so, and nearly a separate story altogether.

They did, you just didn't notice them.

I would argue that this has largely already happened. There is sort of a battle between liberal jewish culture and fundamentalist christian culture for being the official US mythology, but on a day to day basis Japanese culture and values are largely what have filled the power void. I would argue that most of our mainstream institutions, from schools to jobs to pop culture, have become and are increasingly becoming largely Japanese in nature.

In the case of Necromancer though I have to disagree with you. I think that it has held up incredibly well, the parts that were weak when it was released (totally screwy view of technology) are still the weak parts, but the strong parts (the vivid imagery for one) continue to be quite effective.

Then again, I have a strong cyberpunk/modern bias, I'd also recommend a few of Charles Stross's works :)

I personally would add Charlie Stross to this list (http://www.antipope.org/charlie/).

I just started reading David W. Galenson's Old Masters and Young Geniuses: The Two Life Cycles of Artistic Creativity, (http://www.amazon.com/gp/product/0691133808/sr=8-1/qid=13150...), which is where Malcolm Gladwell stole / sourced his New Yorker article "Late Bloomers - Why do we equate genius with precocity?" (http://www.newyorker.com/reporting/2008/10/20/081020fa_fact_...). Galenson says:

There are many common misunderstandings of the history of art, but perhaps none is more basic than the confusion over what determines the quality of art. Although it is of course possible to consider separately the quality of a number of different attributes of an artist's work, the overall importance of art is a function of innovation. Important artists are innovators whose work changes the practices of their successors; important works of art are those that embody these innovations. Artists have made innovations in many areas, including subject matter, composition, scale, materials, and technique. But whatever the nature of an artist's innovation, its importance ultimately depends on the extent of its influence on other artists.

It should immediately be noted that the importance at issue here is not the short run-interest that gains an artist immediate critical or commercial success, but the long-run importance that eventually causes his work to hang in major museums and makes his contribution the subject of study by scholars of art. These two types of success have often coincided, but in many cases they have not. {Galenson "Masters"@2}

There are many possible values a work of art can have, of course, and Galenson is only making clear the set he most plans to consider in his work. Nonetheless, I find his argument intriguing, and I also wonder how many of the works listed in the OP contain some major innovation or innovations then used by other science fiction writers.

On a totally unrelated note, it'd be nice to have HN hyperlinks so one doesn't have to put URLs in parentheses.

I think this misses a huge part of the picture that Gibson, Dick and and other Cyberpunk authors painted (where it differed from other fiction):

1) (Asian) Corporate controlled dystopia - ie breakdown of western-government-controlled world order

2) Information/Data as valuable asset, even commodity

3) Rise of AI and non-human agents

4) Drugs/computing grantings paranormal/heightened state

I'm sure a number of these might have been present in other genres, but these are the hallmarks of the genre as I can see it.

One of the best under rated in terms of "hard" and in terms of quality ideas for me has been Greg Egan (especially his short stories). http://en.wikipedia.org/wiki/Greg_Egan

This is a very weird statement to make, the architects of Cyberpunk (Gibson, Sterling etc.) are Japanophiles if anything, the opposite of Xenophobic. Do you have any evidence to back this statement up?

Cyberpunk was just an extrapolation of the tech scene of the 80s, Japanese tech was massively superior to other countries and, rightly thinking that tech would rule the future, they made Japan the dominant superpower - unfortunately the Lost Decade put paid to that.

Neuromancer isn't just pioneering (which it was), it's incredibly well-written. But like other commenters, I disagree that it's xenophobic, and I don't think its success is rooted in xenophobia in the audience. I also disagree that Gibson's work portrays a "takeover" of Western culture; the ethic ruling the world of Gibson's novels is the ethic of the international mega-corporation, which is hardly a Japanese thing. And also there are lots of non-asian cultural bits: the Turing cops are Swiss, there's the whole cultural slant of Count Zero (not to spoil it for any who haven't read it), the Tessier-Ashpools are French/Australian, etc.

Props for mentioning Cherryh, though. Cyteen is frigging fantastic. Indeed, Cherryh in general is fantastic, although some of her stuff is serious business and some of it is (good) fluff.

As others have commented, hard SF is so very very not dead.

Stranger in a Strange Land is perhaps the purest example of the former, because it's very meta: the thing Heinlein wanted you to think about is the very idea of preconceived notions and full understanding. He spends the first half of the book explaining that you can understand something intellectually without understanding it emotionally, and the last half systematically questioning more and more ingrained ideas (at least at the time) so that you think about them intellectually.

Most of the list is the other type. I'll pick on Dune, since it happens to be number one on the list. Herbert introduces you to a universe that's really completely unlike any other conceived: people fly around in spaceships and have enormous mental powers, but there are no computers (they are illegal), people still fight with bladed weapons (all others are rendered either useless or too dangerous), and universal government has become feudal in nature (for various reasons). People send messengers. A bioengineering project and a social engineering project (gray projects by the same faction, no less) collide, and turn the universe upside down.

As far as hard SF goes, I'd originally clustered that with the world-building type, basically because every good example I've read happens to be in that category. Now that I think of it, though, it's not strictly required. It's really just a question of how much effort they put into reasoning out the science.

This is way off base, and really does a disservice to those authors. Cyberpunk in the 80s may have inherited some economic xenophobia but there were a lot of other ideas in there that I think were more important and continue to be, around the internet and technology.

It also hasn't waned the way you describe. The Deus Ex games are examples of cyberpunk that doesn't have as much to do with Asian takeovers of Western society as Neuromancer or Snow Crash perhaps does. Aside from the technological themes, the demons are thoroughly all-American: media propaganda, military contractors, corporate regulation, big scary government, war on terror, police actions... I would argue a lot of that came from the dystopias envisioned by sci-fi in the 80s, some of it from the cyberpunk authors.

I agree about Neuromancer being fairly average in quality, and that CJ Cherryh should be on the list.

I highly recommend reading Tau Zero by Poul Anderson and Stone by Adam Roberts.

http://www.multivax.com/last_question.html

My only gripe with Chasm City is that there is an (not essential to story) interlude which breaks hard sci-fi canon (ie, superluminal travel).

Try Wolfe's The Fifth Head of Cerberus and see if you might change your mind. It's very readable, and it's pleasantly short. It's half Borges, half Sturgeon, with a pinch of Ray Bradbury.

One author that I find sadly missing from the list is John Brunner. Each of "Stand on Zanzibar", "The Sheep Look Up", and "Shockwave Rider" would arguably deserve a place on this list.

FWIW, I loved Slaughterhouse Five, and had nightmares about it.

There are some excerpts translated by fans: http://dukaj.pl/English/ReadingRoom/BlackOceans

I am trying to think of another context in which your sentence (In <genre>, the sole purpose of <noun> is to mocked, criticized, or denied.) would make any kind of sense whatsoever.

Is there a particular, articulable reason that this is anything more than running away from concepts that challenge your preexisting mindset?

For example, I disagree with slavery, but I also disagree that in modern art, the sole purpose of slavery is to be mocked, criticized, or denied.

I am a fan of Python and not so much of Java. Nevertheless, I would never agree that in web development, the sole purpose of Java is to be mocked, criticized, or denied.

I tend to disagree with the Republican party on a number of issues, but I would never say that in political discourse, the sole purpose of Republicans is to be mocked, criticized, or denied.

Finally, I spend all day, every day working against patent trolls. They make me frustrated and angry, and are, in my opinion, the symptoms of a broken system. Yet I seek out people who advocate for the current system so that I can understand it. I can't afford to have the patent literature I read circumscribed so that the sole purpose of patent trolls is to mocked, criticized, or denied.

Am I missing the point?

As a normative statement, that would be debatable.