I have some ideas, but I'd like my fictional Roman (100BC) General Mikey (ha) to build a submarine(s). This should be able to navigate in lakes and calm shores of the Mediterranean, succeed in bringing down simple boats loaded with soldiers, and finally, travel longer than "just as much air is inside.*" I do not want it to be detected.

Unsure of the number of crew, but just needs to succeed in the above mission. How can I design this advanced machinery, knowing what we know now, but with 100 BC technology?

*) It can surface from time-to-time, but I'd prefer to push the boundary beyond "just as much air already inside".

  • 3
    $\begingroup$ Why not simply look up details of the early submarines? IIRC, there was an attempt to use one in the American Revolution: en.wikipedia.org/wiki/Turtle_(submersible) $\endgroup$
    – jamesqf
    Commented Jun 18, 2015 at 17:06
  • 2
    $\begingroup$ What requirements? You've said "knowing what we know now, but with 100 BC technology". You haven't said anything about requiring a 100 BC mind-set. $\endgroup$ Commented Jun 18, 2015 at 18:53
  • 1
    $\begingroup$ I look forward to reading about General Michaeus Jesterus Comedius. $\endgroup$
    – corsiKa
    Commented Jun 19, 2015 at 14:40
  • 1
    $\begingroup$ Why would you need a submarine - technology had by no means evolved to need such a thing to avoid your ship being sunk by the defenders and there was no explosive that would blow up a ship if you did. Get your own ships and ram the other guy. $\endgroup$
    – Oldcat
    Commented Jun 19, 2015 at 16:52
  • 1
    $\begingroup$ I think the biggest problem is weapons. Submersibles (including diving bells) were known to the ancients, but it wasn't until the late 1800s that that they had reliable weaponry they could use (self-propelled torpedo). The next big problem is propulsion, before diesel/electric submarines long distances weren't practical. Then detectability -- if the water is clear enough to navigate from under the water, anyone can see them, if it isn't clear they need a periscope, which is going to be visible. The effort to build one sub might be better spent building cannon. $\endgroup$ Commented Jun 19, 2015 at 20:22

7 Answers 7


There are two historical examples.

First, the Greeks of Alexander's time created a diving bell style submarine(They also invented a steam engine hint hint). So this is most assuredly repeatable. Modify the diving bell style so the air hose is attached to a flotation device and is dragged along behind, and the soldiers could march along the bottom. (Admittedly, in some areas the floor of the sea would not be solid enough to march upon)

Secondly there is a historical report of a submarine created to travel rivers during medieval Europe. The book (Singular account) reported the inflated pigs bladders for ballast and had a crew of 6-12 men working oars. When the air became hard to breath the Monarch's alchemist released a few drops of something into the air that refreshed it.

Releasing a chemical that steals the carbon from CO2 and releases the O2 back into the contained atmosphere would achieve the same result.

Here is a link. http://www.submarine-history.com/NOVAone.htm

  • 6
    $\begingroup$ @steveverrill Understanding is not really required. Even 100BC people had lots of experience with mining and bad air. Accidentally noticing a specific chemical makes the air easier to breath is not that far fetched. Alchemists were naturally experimenting with weird chemicals in rooms with bad air too. The text sounds like a solution of sodium hydroxide, though. I think it was known in medieval times. And its reaction with acidic oxides would be easy to spot for an alchemist. $\endgroup$ Commented Jun 19, 2015 at 0:34
  • 9
    $\begingroup$ @JorgeAldo NO NO NO AND NO. A Scrubber does NOT replenish O2, they merely remove CO2 from the atmosphere, which is what kills you (not lack of oxygen). $\endgroup$
    – Aron
    Commented Jun 19, 2015 at 9:10
  • 10
    $\begingroup$ @Aron NO NO NO and NO will make you laugh a lot. $\endgroup$
    – corsiKa
    Commented Jun 19, 2015 at 14:42
  • 4
    $\begingroup$ @corsiKa NO, it won't. It will most likely kill you. N₂O on the other hand... $\endgroup$
    – nitro2k01
    Commented Jun 19, 2015 at 18:53
  • 8
    $\begingroup$ I would like to point out that the principle of a diving bell is quite fundamentally incompatible with an air hose connecting it to the surface. For this to work you would need a compressor that pressurizes the air enough to prevent the diving bell from running full of water. $\endgroup$
    – Emil
    Commented Jun 19, 2015 at 18:57

What you want to know is if ancient Romans could design and build an artificial atmosphere. I'm going to say it's unlikely. However, building of the submarine hull and propulsion systems would have been possible. They had access to the Archimedes screw, and to water and air tight wood working. They could have built a shallow water submarine.

Building a compressed air tank to hold extra air probably wouldn't take too much though I doubt the Romans had the kind of precision machining required to make good fittings. This guy has more to say about their limitations. They could have made animals skin bags and compressed some air into those, so no machining needed for those.

The problem in artificial atmospheres isn't getting enough oxygen but having too much CO2. Carbon dioxide scrubbers are used practically everywhere one needs an artificial atmosphere. Many run on soda lime or similar chemicals. Discovery of the properties of soda lime is relatively recent. There's a patent awarded in 1949 to the application of soda lime as a CO2 scrubber. It references a patent for packing "caustic alkali" from 1888. This strongly implies to me that inventing any kind of CO2 scrubber is far beyond what the Romans could have come up with. Sure, they could have gotten lucky but they would have been exceedingly lucky.

Without scrubbers, all those early submariners with their compressed air bottles would know is that their noses burn really bad and they feel a desperate need to get to the surface.

Addendum: The Romans were fantastic engineers. I found this rather long list of technical innovations they made. Too late for the time frame specified in this question but still impressive, the afore mentioned list indicates that by the end of the 3rd centry AD, Roman engineers had all the mechanisms required to make a steam engine. They didn't but all the components were there.

  • 3
    $\begingroup$ The Romans had limekilns and made a lot of lime for building cement. Whether they could have recognised this additional use for it is another matter though. $\endgroup$ Commented Jun 18, 2015 at 20:42
  • $\begingroup$ @BrianDrummond, that tidbit of info, I did not know. Thank you. :) $\endgroup$
    – Green
    Commented Jun 18, 2015 at 20:44
  • $\begingroup$ @BrianDrummond yes, plausible! See my post. $\endgroup$
    – JDługosz
    Commented Jun 19, 2015 at 2:33

Submarines for war (especially in a world without sonar or helicopters) need not be fully submersible. They can be semi-submersible. That is, the body of the ship is just under the surface of the water and parts of the ship such as snorkels and periscopes are always above the surface.

Because they're never out of contact with the surface they don't need air supply, just good ventilation. They can be very low tech. Only the knowledge of the physics of floating (Archimedes) is needed to build one.

Today, semi-submersibles are better known as cocaine submarines or narco submarines. Most are built in Colombia and are used to smuggle drugs (google "narco submarine"). Some of the more advanced versions can be submersed for short periods of time.

  • 1
    $\begingroup$ Oh hello, NSA ... :) Remember that Google doesn't forget what you search for. $\endgroup$
    – aikeru
    Commented Jun 19, 2015 at 20:34
  • $\begingroup$ Good they can have 50,000 red herrings to wade through. $\endgroup$
    – Joshua
    Commented Jun 20, 2015 at 4:59
  • $\begingroup$ I agree, but I'm worried about them disrupting the surface, as well. Great answer. $\endgroup$
    – Mikey
    Commented Jun 20, 2015 at 6:18

How to build a submarine in 100BC with materials and technologies known to the Romans at that time based on this list of technologies. We assume a modern naval architect finds himself on the shores of the Mediterranean

Weapons systems: Brass drills or reciprocating saws mounted to the hull of the submarine used to attack the keel of enemy ships. Ship's carpenters can't fix broken keels at sea but they can fix holes in the hull. The submarine lacks the kinetic energy/momentum for ramming attacks.

Hull: Easy. Peoples of the Mediterranean have sailed the seas for centuries/millenia before our naval architect appears. Forming wood into air/watertight shapes shouldn't be difficult for an accomplished shipwright. This part is definitely solved.

Propulsion: A primitive screw could be cast from brass. Working with cast brass opens up huge possibilities for structural components in this submarine.

Visibility: The Romans were surprisingly good glass blowers. Our architect may have just needed to order some pieces of specific dimensions to form portholes in the hull.

Atmosphere: This is the tricky part. As our architect is not a chemist, he lacks the chemistry knowledge to make soda lime (or any of the other CO2 scrubbers described in other answers.) He would know about compressed air and may be able to construct a primitive air compressor stored in metal tanks. Seals could be made from oiled leather or wax compressed between two pieces of metal to make the air tank hold air better. Pipe could be made though this might be a trade-off on engineering time.

Psychological Warfare: He knows that the Romans and Roman enemies are superstitious people. So he would add some elements to the submarine that if seen while surfaced would make anyone who saw it think they had seen a sea monster.

Artisan Availability and Funding: Many of these innovations would require the best artificers that the Roman empire had to offer. That many people and the quantities of materials for experiments would require large funding requirements and a powerful sponsor. Getting into this kind of a position is a completely different question. Perhaps some inspiration from Twain's "A Connecticut Yankee in King Arthur's Court".


I do not believe this is possible:

TL; DR: The submarine will be far, far, too slow and cannot easily be weaponised by a Roman.

A diving suit with Soda Lime CO2 extractor might work, but I have not covered that.

Water pressure

The Romans certainly could build ships. It makes sense that a Roman shipwright could build a vaguely submarine-ish, underwater vessel. Since the ship will be underwater, however, it will have to withstand a slightly higher water pressure. Wolfram Alpha tells me that the pressure 10 feet below sea level is 130kPa. As Mike L points out in the comments, a wooden hull would bend and flex underwater, just like a regular ships hull. As the hull bends seams will briefly open and the submarine will leak. On an ordinary ship, this water is pumped or bucketed overboard, but you can't really do this on a submarine. The sub would sink and not be able to surface. You could make the submarine out of a "watertight all-metal rigid hull" I doubt the Romans could easily cast a hollow submarine sized piece of metal. Making the ship out of smaller metal plates would introduce the same problems as a wooden ship.


Another problem would be manoeuvring. Modern submarines dive by taking water into special compartments and surface by removing the water from those compartments (usually by pumping in a gas). This would be difficult for the Romans to replicate because they did not have the sort of precision engineering required to make such a device, nor do they have access to pressurised gas to flush the dive compartments.

Overcoming Friction

Disclaimer: this section is entirely based on my 10 minute google. If you (the reader) know what your talking about then I have no problem with you improving this with an edit. Hint hint.

One thing the other answers do not cover, is that moving underwater is hard. This physics.SE question and this linked site helpfully give the following equation for finding the drag force exerted on your submarine:

$$F=\frac{1}{2} ρv^2A c_d$$

We will assume that the front of the submarine is spherical (assume a spherical submarine in a perfect vacuum...) with a radius of 2.5 meters. If I understand the table from the second link correctly, then the drag coefficient ($c_d$) of the hull should be about $1.1$ (or between a human and bundle of wires - which is my (very bad) estimation for a wooden sub).

We have:
$c_d = 1.1$
$ρ = 1000$ (density of fluid, $1000kg/m^3$ for water)
$v = 3.3$ (Target velocity, $m/s$, ~ 2 times walking speed)
$A = 2\pi2.5^2 = ~39.27m^2$ (Area of the front of submarine)

Which gives us:

$$ F=\frac{1}{2} 1000 \times 1.1^2 \times 39.27 \times 1.1 $$
Or about $104536$ Newtons of resistance to travel at walking speed, which is insane. A single human can push around 630N with a firm surface to push off, on land.

Even if my calculation is outrageously wrong (it must be), I do not believe a single Roman or small group of people would be able to propel the submarine at any speed.

Roman ships had many oars. The trireme for example had 170 oarsmen. If we put all 170 in the submarine described above, they would have to exert 615 N of force each to maintain a speed of $2.2 m/s$ ie: slightly more than walking speed.


With 170 men aboard, your ship will need lots of air. I do not believe that the Romans could have compressed air, at least not to the pressures required. Soda Lime might be enough of a few people: (Quote from my early draft)

As the other answers have pointed out, removing CO2 from the ship is your primary concern and you can do this with Soda Lime, which is made from Calcium hydroxide (limewater, which the Romans had access to) and Sodium hydroxide (aka caustic soda, aka lye). Making Soda Lime is well within the Roman's grasp.

But I realised, that's just not going to cut it with 170 men. You will either need a funnel and bellows for air (better be a big funnel) or a lot of Soda Lime.


The trireme relied on speed to ram enemy ships becaue the Romans did not have cannons or other convenient point and shoot weapons. Some ships had trebuchets. Ramming is out of the question for the sub, as are underwater trebuchets, so weaponisation will be difficult if not impossible for your Roman general. Dronz points out in the comments that an underwater ballista could be used, point blank. This is possible but the performance of the ballista will be affected by water, so the final speed and power of each bolt would be limited. Weaponising the sub will still be difficult.

  • $\begingroup$ How about an underwater ballista-like mechanism, used point blank on bottom of target hull? $\endgroup$
    – Dronz
    Commented Jun 19, 2015 at 23:12
  • 1
    $\begingroup$ An important drawback that should be added to what you said: Wooden ships flex and bend under the differential pressure cause by the sea. In the process, seams (briefly) open and the ships start to leak; this can't be helped without making a watertight all-metal rigid hull. A wooden submarine that's watertight on the surface will quickly start leaking once it submerges, and you can't just dump (nor pump) the water overboard. Since it now has negative buoyancy, your submarines will start disappearing underwater and you'll have no idea why. $\endgroup$
    – Mike L.
    Commented Jun 20, 2015 at 1:14
  • $\begingroup$ @MikeL. That is some good information (which I have added). I don't think the Romans would have been capable of building an all metal hull, however. $\endgroup$
    – amziraro
    Commented Jun 20, 2015 at 5:32
  • $\begingroup$ You're implying that human powered submerged propulsion is not possible. But early 19th century submarines were indeed hand-cranked by a crew. How did they do it? $\endgroup$
    – JDługosz
    Commented Jun 20, 2015 at 7:14
  • $\begingroup$ @JDługosz They had the advantage of ~1800 years of more advanced engineering and the industrial revolution. The Romans did not have metal hulls, which could make smaller and lighter submarines. $\endgroup$
    – amziraro
    Commented Jun 20, 2015 at 10:08

Also look at Leonardo's diving system (see also here).

While working in Venice, the "water city", in 1500, da Vinci designed his scuba gear for sneak attacks on enemy ships from underwater. The leather diving suit was equipped with a bag-like mask that went over the diver’s head. Attached to the mask around the nose area were two cane tubes that led up to a cork diving bell floating on the surface.

Air was provided from the opening of the tubes to the diver below. The mask also was equipped with a valve-operated balloon that could be inflated or deflated, so the diver could more easily surface or sink. Additionally, Leonardo da Vinci’s scuba gear invention incorporated a pouch for the diver to urinate in.

diving app The TV show I saw where they built and tried it (correcting a deliberate flaw in drawing) found that the floating part would hold a resavoir of air when pulled completely underwater. Soldiers could move while snorkeling and go full-stealth for final approach, or dive deep for whatever work they were doing. (see this video timecode 3:50 for the "pull under" concept I'm talking about.)

That could be made by Romans and used as infantry. The float/resovoir evolves into extended "all under" time using larger air bladders, and shared snorkel buoys that turn into small watercraft themselves.

Picture the interesting difference from your standard sealed-hull submarine concept: divers spreading out, or collected in an open framework to and from the site, rather than an enclosed can.

As this technology would certainly be used for building projects too (as casons and bells were eventually in our civilization), it is plausible that they might figure out that carrying lime makes the air stay good longer. The army was also the engineering corps, and would be building dams and harbors as well as wrecking the enemy's. Being familiar with the underwater equipment, they might naturally try using something like casons or bells for efficient construction work, to stay at the worksite longer: it's just the pull-under resovour supersized and suken using ropes and weights. The evolved concept of a non-sealed craft decendent from the simgle-user float/bag is essentially a moble diving bell.

They would carry cement and bricks and such in the bell, I would think, since that is what they are using it for. And they might find that cement (quicklime is the main ingredient) makes the air last longer.

The only component that would not be avilable in 100BC is the crystal glass window. Without introducing a source of quality clear glass, you would need to figure something else. Maybe they found tiny transparent pieces of glass or quartz mineral and just made peep-holes, or didn't cover the whole head but used a mouthpiece like today's snorkel.

  • 1
    $\begingroup$ If you try to mix a snorkel with a diving bell, the result isn't pretty. If the snorkel is more than about a foot long, human lungs can't pull air in against the water pressure; a few more feet, and you start risking crush injuries as water pressure tries to force the user into the snorkel. $\endgroup$
    – Mark
    Commented Jun 19, 2015 at 3:53
  • $\begingroup$ I saw a show where people built Leonardo's design (correcting the intentional misinformation) and showed how it could be pulled under keeping a reserve of air. Yes, its hard to inhale when walking in battle gear to board the ship, not much problem if swimming stealthily just under the surface. The bell is not connected to the surface: the bouy becomes a bell when pulled under. Also, mentioned using bells separately from that as a natural thing they would use for construction once familiar with diving aparatus. $\endgroup$
    – JDługosz
    Commented Jun 19, 2015 at 7:29
  • $\begingroup$ Why a downvote? It's a perfectly good exploration of the subject, how it might evolve, and considers how and why quicklime might be discovered for CO2 scrubbing. $\endgroup$
    – JDługosz
    Commented Jun 19, 2015 at 7:31
  • $\begingroup$ @Mark: Note that in "scuba mode" when pulled underwater the water pressure would also act on the reservoir of air therefore the pressure differential issue you mention shouldn't be a problem. $\endgroup$
    – slebetman
    Commented Jun 19, 2015 at 8:43

Romans where able to cast large metal structures. You CAN cast big metal pieces on sand, by digging a hole into the ground and using a piece of wax as mold. IF trully needed this part is not that hard. A big bronze bell is not too different from this. But, the problem with tech is that you do what you believe is at your reach. Talking to a roman about building an underwater ship is like asking a medieval man to take a trip to hell. He wont accept it.

You can use a big closed bell to go underwater. As you dont have compressed air, you cannot use ballast tanks to control depth. You have two options : Tie ballast weights to your hull via cables that you can control the length. When the ballast weights hit the ground your submarine will stop diving (the bell being less dense than the displaced water), give it lenght and your submarine climbs etc. The other option is to keep two floats over water and tie your submarine to them. The first option is risky, you dont know how deep the ocean is (yet), the other option will keep a target visible all the time. If the floats are made from bronze or other strong metal, and you make your attack at night and they are painted black, your target might not have a way to detect its approach.

The most realistic approach, is to use divers and attack ships in harbour.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .