Science Fiction has always been an inspiration for real life science accomplishment. Robert H. Goddard, one of the fathers of rocketry, was inspired by H.G. Wells’ “War of the Worlds.”
And perhaps as soon as the first modern rocket took to the sky with a thunderous woosh, the imagination of writers and engineers turned toward reversing that ascent and landing the beast back on its flaming tail.
The first depiction I remember of a rocket standing upright on its tail fins after a controlled descent is from Robert A. Heinlein’s 1947 classic “Rocket Ship Galileo,” which also happens to be the first book I ever read.
As I grew up in the late 50’s and early 60’s, my memory is filled with scenes from TV and movies of silver bodied variations on the German V-2 rocket landing tail first on the moon or Mars. “Destination Moon,” “Rocketship X-M” and “It! The Terror from Beyond Space” all featured iconic silver needle designs that could plant the perfect landing after an exciting display of orbital athletics.
The first real life ascent and soft descent of a rocket may surprise you, because there were no silver-bodied spaceships involved. In 1961, Bell Aerosystems demonstrated the Bell Rocket Belt, a wearable rocket pack able to lift a human to about 60 feet, give them thrust and maneuvering control, and the ability to descend safely at the end of an exciting 21 second flight.
The Bell Rocket Belt proved popular at demonstrations and was seen on screen in the likes of Thunderball, Lost in Space, and Ark II. Unfortunately, the device’s severely limited flight time made it impractical for continued development.
Technically, the moon landings were controlled rocket descents, but they had only 1/6th Earth’s gravity to contend with and some unique, retractable spidery legs to plop down on.
If human feet and spider legs don’t count for a tail first descent, you must wait 30 years until the DC-X, a true rocket, came under design and development at McDonnell Douglas in 1991. The DC-X was part of the Strategic Defense Initiative, conceived by Maxwell Hunter, a prominent aerospace engineer, and strangely enough, championed by SF writer Jerry Pournelle, who helped set up a meeting with then Vice President Dan Quayle.
The DC-X was sold as a system that would deliver cheaper, reliable rocket flight with fast turnaround times. A prototype was built, but after 3 test flights and achieving a maximum altitude of 2.5 kilometers, the project ran afoul of funding cuts. NASA revived the idea for a while in the mid-1990s, but again funding for the DC-X was eventually diverted toward a Lockheed Martin design for a Shuttle replacement, and the DC-X experiment came to an end. 
While not a practical workhorse, the prototyping of the DC-X paved the way for more successful, this time commercial, ventures in the form of Blue Origin’s New Shepherd and the SpaceX Falcon 9. In fact, Blue Origin benefited by hiring a number of DC-X engineers.
Why Is This So Hard?
To achieve a tail-first landing, a rocket must have enough thrust to take off and accomplish its mission while retaining sufficient fuel to descend. It will have to relight its engines, possibly at hypersonic speeds, then gimble (tilt) its engines to direct thrust at just the right angle, and finally throttle those engines down at the perfect rate to plant a stable landing.
All of this takes microsecond timing and is likely beyond the ability of a human pilot relying on analog controls and mechanical gyroscopes. It had to wait until the age of fast, robust computers exchanging data with reliable, hardened sensors and communicating with GPS and other positioning systems. 
And even then, nothing related to this history altering feat is simple. SpaceX was teetering toward failure after many partially successful tests when Flight 20 accomplished an orbital launch mission and brought its Falcon 9 booster back to the pad, landing on its tail, December 21, 2015.
Since then, Falcon 9 boosters have been landed and reused more than 100 times. The SpaceX Starship and Blue Origin’s New Shepherd booster have also demonstrated the modern reliability of tail-first landings.
It’s All About the Launch Cost
After the Apollo program, the cost to launch a kilogram into low earth orbit held steady at around $18.5K/kilogram for decades.
Put into service in 1981, the “reusable” space shuttle did something interesting. It cost more at first. Even so, launch costs over the shuttle’s lifetime went from $85K/kilogram to $27K/kilogram.
In 2008, the SpaceX Falcon 1 brought launch costs below $10K/kilogram.
By 2017, Falcon 9 (the reusable one), could launch for less than $2K/kilogram.
Falcon Heavy launches material into low Earth orbit for under $1K/kilogram.
NASA’s stated goal is to reduce launch costs to “tens of dollars per kilogram” by 2040. But we don’t have to rely on NASA anymore, and the world may proceed faster than it’s possible for government agencies to adapt.
Based on the reusability of rockets, following in the imaginative footsteps of those 1950s science fiction authors and cinematographers, the space race is back on. This time, though, companies like SpaceX, Blue Origin, and Virgin Galactic are vying for pride of place in building the first commercial space lanes.
Today, the world’s first purpose built commercial spaceport is taking shape in New Mexico with the vision of becoming the premier multi-modal Spaceport. 
And at Starbase Texas (Boca Chica), the Starship program is marching step by step toward the first manned mission to Mars. Even before then, Starship’s tail first descent and landing has been chosen by NASA as the vehicle to return us to the moon before this decade is out.
And when it does, the science fiction vision I saw so often growing up will be more than the stuff that dreams are made of.