I don't know when I'll get the time to reread Frank Herbert's original DUNE and thus do an "official" Reading Rheum review of it. But since I have read the book three times, I have a reasonably good recollection of its major tropes and conceptual scope. My main aim here is to set down some general ideas about the novel so that I don't repeat myself when I cover the David Lynch film on my movie blog.
I've also read a pretty fair sampling of Herbert's other science-fiction novels, and though I've not looked at any of them in the last twenty years, my overall recollection is that none of them exhibit the mythic imagination of DUNE. But most of them follow the pattern of good didactic science fiction: they set up some intellectual problem, based on some metaphenomenon predicated upon sci-fi's famous "one gimme" rule, and proceed to discuss the societal or psychological ramifications resulting from the phenomenon. But there's usually not a lot of symbolic depth in purely didactic arguments, though, as I've argued frequently on this blog, sometimes the didactic and mythopoeic forms of discourse can work together to good effect. But this didn't happen with most Frank Herbert books, which are mostly concerned with didacticism-- much like the majority of the DUNE sequels, though I admit I've not reread any of these in twenty years either. I haven't taken any surveys of science fiction fandom, but the dominant impression I've gained from both personal conversation and message boards is that almost no one likes any of the sequels better than Original DUNE. One can find fans who like BRIDE OF FRANKENSTEIN a great deal more than the 1931 FRANKENSTEIN, or THE EMPIRE STRIKES BACK more than the 1977 STAR WARS. But in a statistically dominant sense, DUNE is the "first child" that everyone likes, and all the sequels are the equivalent of red-headed stepchildren. My perhaps-superficial, definitely-not-researched impression is that when Herbert conceived DUNE he drew upon a number of intellectual interests-- the ecology of sand-dunes, the effects of psychotropics on human perception, and perhaps most of all, the mystique of the savior-- and allowed himself to build on all of these concepts in a manner more mythopoeic than didactic. The didactic impulse was certainly there, though. Herbert stated in interviews that he set up the Campbellian heroic structure of DUNE with the long-range intention of undermining the savior-mythology behind the rise of the heroic Paul Atreides. This authorial intent is particularly strong in GOD EMPEROR OF DUNE, where the great Messiah of the Spice mutates into something akin to a sandworm. And yet, Herbert sold his myth-world so thoroughly that most readers were as immersed in that world as Herbert himself was when he created it. The author created a beautiful dream just so that he'd be able to wake the dreamers from their illusion and reveal, "see, you shouldn't have fallen for my glamorous hoax." Instead, many if not most readers saw Herbert's deconstruction of his original dream as the real illusion-- again, judging purely by the general fannish opinion that the later books were inferior to the original. (I will add that I found a few of the early sequels at least interesting in their own right, including GOD EMPEROR, but some of the later books are entirely forgettable.)
Speaking as I was of influences, some critic, whose name I did not preserve, remarked that the sandy wasteland of the planet Dune had an interesting predecessor in science fiction literature: "the sands of Mars," as Arthur C Clarke called them. And the foremost mythographer of Mars in early science fiction also dealt in a lot of the same elements of combative adventure and medieval intrigue as Herbert: the redoubtable Edgar Rice Burroughs. Now, the Mars books of Burroughs are as bereft of didactic insights as the majority of Herbert books are lacking in mythopoeic power. The unremembered critic argued that Herbert had to some extent built upon Burroughs' high-adventure mythos and imbued it with far greater subtlety and intellectual heft, and I agree that this is certainly possible, even if Herbert only knew the John Carter series by reputation. But I'd argue that there's another Burroughs series that may have had more structural impact on Herbert, and that's the Tarzan series, which, more than the Mars books, Herbert could have known from cinema had he never cracked one of the original stories. The trope common to both Tarzan novels and Tarzan movies might be boiled down to "good colonists fighting bad colonists for the control of tribal resources." Tarzan, the scion of good colonists, doesn't "go native" like various Joseph Conrad protagonists, but rather "goes ape," which contingency makes the ape-man into a superhuman figure. (A drug-free one, by the way.) The morality of Tarzan's interactions with Black Africans, corrupt Europeans and ruthless Arab slavers was not something Burroughs could have addressed intellectually, even had he wished to do so. But when in DUNE we see two great Renaissance-style families vying to take control of Planet Dune's spice-commodity-- the noble family Atreides and the decadent Harkonnens-- what we're seeing is an old wine decanted into a new, and perhaps more elaborate, bottle. It's to Herbert's credit that in the early novels he doesn't ever reduce the entire three-way struggle to pure politics-- though I can't speak to the later ones.