Writing Good Military Science Fiction

(Updated 13 March 2014)

All military units, tactics and strategies are products of their time.

While there are a few global permanent operating factors at play that work all the way from Ancient Rome to 2300 AD; a lot of things change.

What works for a SEAL Team in 1970 Vietnam won’t work for a SEAL Team in 2048 Somalia!

This is a serious problem with a lot of science fiction authors – they “file off” the feel of a particular era or time – say, the Vietnam War – and transpose it forward into their fictional universe, without taking into account the factors in play that made the Vietnam War the Vietnam War as we know it.

Technology is the single biggest factor driving change.

Vietnam turned out the way it was because of the technologies extant then at high readiness levels for large scale military use by both U.S. and North Vietnamese forces.

Case in point – a lot of Vietnam was covered by heavy triple-layer canopy jungle; and with 1960s levels of technology, light infantry or guerillas had the advantage there, in that they could move units or supplies relatively undetected through the jungles – e.g. the Ho Chi Minh trail which supplied NVA Main Force units or Vietcong cadres in South Vietnam.

There were attempts by the US to use technology to strip away the cover of the jungle, such as using AC-130s equipped with advanced technologies (for the time), such as early FLIRs, radio direction finders that could home in on the electrical sparks from the primitive unshielded engine sparkplugs used by North Vietnamese trucks on the trail and the first unmanned sensor networks (IGLOO WHITE, which cost about $6 billion in 1960s dollars) deployed on a wide scale.

In the end though, everything was still primitive enough to require large scale infantry operations to clear an area of guerillas via the expedient of walking around the ground and poking things. Even then, large amounts of manpower on the ground wasn’t a guarantee of success – the tunnel complexes around Cu Chi remained operational, despite repeated large scale infantry operations to clear the “Iron Triangle” out.

There are several main factors driving major changes in military tactics in recent years; and they are:

Sensor Capability is Increasing Every Day

Today, the US military is testing out advanced radars on drones which can penetrate triple-layer jungle and see movement underneath via the advancement of cheap, lightweight computers being able to do a lot of signals processing voodoo on the raw radar data to filter out vegetation noise without destroying too much of the data quality through false positives.

Combine that with sensor fusion – being able to combine data from disparate sensors (Infrared, Visible Light, Ultraviolet, Radar) and it becomes really hard to conceal something from an enemy actively searching for it. You may be able to hide from one sensor, but you can’t hide from them all.

For example, towards the end of World War II, they began to use infrared photographic film on reconnaissance aircraft to detect ‘dead’ vegetative material, which indicated that something had been camouflaged:

Uncamouflaged Factory Complex
(Visual Light Film)

Camouflaged Factory Complex
(Visual Light Film)

Camouflaged Factory Complex
(Infrared Light Film)

(Live vegetation is white,
dead vegetation is black)

The camouflage that worked just a few short years ago in 1940 no longer worked in 1945.

Another factor to consider is automation. Aerial reconnaissance for D-Day required 1,700 officers and men to study 85,000 photo-recon images daily in the days leading up to and following the invasion. Imagine the level of accuracy you could get now with modern computers automating much of the drudgery.

Actually, you don’t need to.

Around 2007, the US Army began to deploy the Buckeye Precision Geo-Referenced Digital Airborne Camera System on manned aircraft which flew up and down heavily contested ground routes on a regular basis.

The BUCKEYE system created detailed digital 3-D pictures of the streets below, allowing people to “drive through” them on a computer, and allowed automated image analysis to spotlight the changes between imagery taken a few hours apart from each other. Thus, if insurgents dragged a dead cow by the side of the road to use as camouflage for a massive IED, BUCKEYE would spot the no cow / cow changed status and flag it for human analysis.

Successful IED attacks in some cities dropped as much as 90% after BUCKEYE was used there.

Sensors are also becoming more and more abundant and available to ever smaller units of employment.

Take for example, acoustic sensors. They’ve been used in Iraq and Afghanistan to spot snipers from their gunfire; and in several major US cities (Washington DC amongst them) to localize crime by the sound of gunshots.

What happens when every soldier has a series of acoustic sensors on his helmet and the sensors in a platoon are all networked together; so that if a sniper fires at them from the trees; the platoon commander automatically knows within about three seconds where the shot came from?

The role of the sniper then changes dramatically, if each infantry platoon has an automated countersniper system consisting of an orbiting UCAV that fires miniaturized precision missiles at sniper locations within 4 seconds of a shot being fired at the platoon.

Suddenly, the sniper is no longer a highly trained specialist who is a crack shot with a rifle, but more of an ambush predator who sets up scores of unmanned autonomously targeting systems that can be triggered when he is very, very far away.

Likewise, stealth weaponry changes. During the Vietnam War, US Special Forces routinely went out on operations using suppressed Browning Hi-Powers nicknamed “Hush Puppies.” They were used to silence sentries and barking dogs. What happens when enemy camps have acoustic gunshot sensors (suppressed weapons still make noise) and the guard dogs have little microtransmitters embedded that alert their handlers if the dog is seriously wounded?

Yes, you could try jamming the dog transmitters, but that’d be a dead giveaway that something is up, because any competent dog handler would assume that his dog had been seriously hurt if he suddenly stopped receiving health status updates.

Weapons Precision

During World War II, the following weapons accuracies were achieved:

Aircraft Type


Dive Bombers
Tactical Fighters

98.4 feet (30m) (Ju-87 Stuka, Expert Pilot, 80 degree dive, pullout at 2,300 ft)
175 feet (53m) (SBD Dauntless, Average Pilot, 80 degree dive, pullout at 1,700 ft, c.1944-45)
180 feet (54.8m) (P-47 Thunderbolt, No Flak, steep dive, release at 1,500~ feet)
300 feet (91.4m) (P-47 Thunderbolt, Moderate Flak, steep dive, release at 1,500~ feet)
420 feet (128m) (P-47 Thunderbolt, Heavy Flak, steep dive, release at 1,500~ feet)

Strategic Bombardment

3,400 feet (1,036m) (B-17/24, Day Level Bombardment from 22,000 feet c. January 1943)
1,100 feet (335m) (B-17/24, Day Level Bombardment from 22,000 feet c. 1944-45)
850 feet (259m) (B-29 with APQ-7 radar, Night Level bombardment from 15,000 feet c. 1945)

By Vietnam, advances in miniaturized radars and bomb computers on all levels kept CEPs the same or smaller, despite much higher speeds and release altitudes versus WWII, as shown in the table below:

Aircraft Type


Tactical Fighters
(F-4 / F-105)

323 feet (98.45m) (Daylight, Shallow Glide, Bomb Release at around 6,000 ft)
500+ feet (152.4m) (Daylight, Shallow Glide, Bomb Release at around 9,000 ft)

Tactical Attack

656 feet (200~ m) (F-111A, Night Time, Level Drop, Bomb Release at 500 ft and 480 kts)
60 feet (20m) (A-7D Corsair II, Daytime, Shallow Glide, Bomb Release at Moderate Altitudes)

Strategic Bombardment

Approx 1,000~ feet (304m) (Night Time, Level Drop, Bomb Release at 30,000 feet c. 1972)

In both wars, the lethality of the 500 pound bomb was crudely about 50 feet (15.24m). Lethal radius depends on the target being attacked – the same bomb that could rate a lethal radius of 200 feet (60.96m) against infantry standing in the open could have a lethal radius of just 25 feet (7.62m) against a light armored vehicle.

There were also several major changes between World War II and Vietnam that changed how airpower worked were:

Using some math with the data above, we can work out the following cases for attacking a point target, if we assume the lethal radius of a 500 pound bomb against our target type is 50 feet, and we desire a 85% chance of destroying the target.

From the examples given above, can you see how the changing lethality/accuracy of each era’s weapons system drives the tactics and thus, the “feel” of that era of combat?

World War II, huge bomber streams of hundreds of aircraft, due to the low relative payload of each aircraft and inaccuracy of optical bombsights, and the need for each aircraft to stay within the defensive envelope of a turret mounted fifty caliber machine gun.

Vietnam, small ‘cells’ of 4 to 15 aircraft, due to the higher relative payload of each aircraft, better accuracies, and relying more on electronic jamming for self protection than internal weapons.

Now imagine how a guerilla campaign would fare if the government forces had UAVs that could orbit over a fixed point of ground for hours on end, just watching; and when they found something of interest, could precision strike it with 100 pound smart bombs or missiles aimed at individual enemy personnel?

Actually, we don’t need to imagine this, it’s already happened in Colombia. (Washington Post Article)

Essentially, the Colombian government used wide area surveillance equipment to locate FARC leadership groups in the jungle, then used A-37 Dragonflies armed with Enhanced Paveway IIs, which have dual Laser/GPS guidance to destroy said FARC leadership groups.

Before, it used to be that achieving bombing results of this lethality required fairly large, highly trained specialist units, which meant that in a second or third world air force fighting a counterinsurgency campaign, there would be massive amounts of leaks due to the size of the specialist unit needed and the corruption inherent in those countries.

Now? All they need to do is take a couple of average pilots, compartmentalize them into a small, tightly run group due to smart bombs removing the need for both large unit sizes and extreme training.

Increased Automation

I raised this earlier with the automated analysis of reconnaissance imagery for changed conditions. But unmanned combat systems have the potential to dramatically transform warfare, at least in the second and third worlds.

Traditionally, one of the reasons that insurgencies grind on for so long in the second and third world is that the government militaries are generally very low on the competence scale, because the governmental leadership is afraid of a coup d’etat, which generally are led by elite units – Hugo Chavez in Venezula was a paratrooper, for example.

What happens when combat UAVs advance far enough that any country which can operate a single engined turboprop trainer aircraft can operate a fast jet UCAV capable of carrying at least two 500 pound bombs? Suddenly even the poorest African nation has a deadly competent deep strike capability.

Elsewhere in the military, what’s the impact going to be when effective ground combat robots are introduced and filter down?

Suddenly second/third world militaries now have a pretty reliable, highly trained force that can strike with precision and avoid the problems inherent in using ill trained 18 year old conscripts who tend to shoot first, causing messy situations that exacerbate insurgencies.

A collorary to this is that coup d’etats become easier if all you need to do is gain access to one of the fast jet UCAVs and program it to bomb the presidential palace at 3AM while el Maximum Leader sleeps.

So...be very careful in pushing your Vietnam-War-in-Space theme, people.