Following on from my earlier question here: Can a water transport canal be excavated by water power?

I have abandoned the self channeling as too difficult. So manual digging will be required. I need to get a handle on how big the canal might reasonably be. This will obviously depend (among other things) on the time and manpower available, so to simplify I will talk about cubic metres moved per man day.

Assume the same sand/loam/clay surface but add a solid rock foundation a few metres down (at a depth that is most convenient to construction). Take manual labor using basic picks, shovels and wheel barrows as an initial starting point for cubic metres dug per man day calculation.

The question is how much more efficient than this basic set up can they make their operation if they use a wider range of mechanical and animal based equipment and manage the work intelligently? All pre industrial devices are in scope, levers, pulleys, cranes, wheeled carts, horses, other animals and anything else that you can think of. What level of multiplier should I add to the basic mandrolic method?

There are no steam engines or internal combustion engines, but this society is very advanced in other respects. For example they are familiar with all manner of hydraulic calculations such as the manning equation for channel flow estimation.

Related: How long would it take to dig a 10 km canal with Bronze Age technology?

  • 2
    $\begingroup$ (a) Do we have explosives? But more to the point... (b) What's your question? How much more efficiently (aka, how much more quickly) can the canal be dug? As in, "humanity did X historically, but using hindsight, could we improve on that?" This is something of a weird question. Strainest thou at a gnat? $\endgroup$
    – JBH
    Commented Nov 18, 2020 at 16:54
  • 1
    $\begingroup$ Just look at real world canals, a lot of canals were build before the industrial era. China and Britain being two poster children for it. As it is this question needs a lot of work. Consider just asking how fast can a canal be built preindustrial. The Bridgewater Canal would be a great example. $\endgroup$
    – John
    Commented Nov 18, 2020 at 17:19
  • $\begingroup$ Its true there will be a lot of calculation (which I am happy to carry out) that’s why I tried to simplify the issue by avoiding the canal calculation. I’m interested in the technological advantage in digging of the late preindustrial period technology over very primitive digging methods – what multiplier to apply? x2? X5? $\endgroup$
    – Slarty
    Commented Nov 18, 2020 at 17:32
  • 1
    $\begingroup$ The 240 km (150 miles) Canal du Midi was built in 14 years (1667 to 1681) over mountains and valleys, entirely by hand. And it wasn't even some sort of national effort of the Kingdom of France; it was an ordinary engineering project, one of many. $\endgroup$
    – AlexP
    Commented Nov 18, 2020 at 19:28
  • $\begingroup$ If you're really building a world, why does reality trump realism? Either way, why not ask Google how long this, that or the other interesting canal actually took to build? These days, some even have their own TV doccoes! $\endgroup$ Commented Nov 18, 2020 at 23:08

2 Answers 2


The longest canal in the world is also one of the oldest. The Jing–Hang Grand Canal in China.

The oldest parts of the canal date back to the 5th century BC, but the various sections were first connected during the Sui dynasty (581–618 AD). Dynasties in 1271–1633 significantly restored and rebuilt the canal and altered its route to supply their capital.

The total length of the Grand Canal is 1,776 km (1,104 mi).

This canal was built in part through a mountain range, with significant construction challenges, without augmented power machines. So certainly building your 400 km. canal on relatively flat lands in soil very amenable to digging (with very few engineering challenges) by a non-powered society is well within reasonable conjecture.

And I know from personal experience, having worked my way through University by working in a quarry, that a reasonably fit individual in a safety-conscious highly unionized environment can unassisted manually move six to eight cubic yards of material from an enclosed conveyor belt basement in an eight hour shift, using only a shovel and wheelbarrow, without an unreasonably excessive expenditure of effort. All breaks included. Troglyte labour could easily do it in half the time.

  • $\begingroup$ Thanks for that! I was trying to establish what advantage the late preindustrial tech would give for instance heavy lifting cranes $\endgroup$
    – Slarty
    Commented Nov 18, 2020 at 18:16
  • 2
    $\begingroup$ I can absolutely assure you that digging a trench 5 to 10 meters deep in very forgiving soil and moving the material a very short distance out of the way would not be significantly benefited by cranes and such, providing you have the manpower. Cranes are only useful for heights and really heavy objects (boulders). The classical wheelbarrow and wagon would suffice. The only advantage powerd machines have (and basically they are just powered wheelbarrows, shovels, and wagons) is in manpower. $\endgroup$ Commented Nov 18, 2020 at 18:48
  • 2
    $\begingroup$ Consider the Western Front in The Great War was 700 km. long, and was essentially one long continuous trench, dug essentially by hand, upwards of four to six meters deep and three to four meters wide, and it was manned by one contiguous line of soldiers from one end to the other, completed in only a few months to a year at most. $\endgroup$ Commented Nov 18, 2020 at 19:00
  • 2
    $\begingroup$ @Slarty it gets even better. The romans could already build a 90km aquaduct with water flow incline of 0.3%, dropping 264m in height. Most of it is until the halfway point, dropping 0.15% the rest of the way. Although aqueducts are different from canals, they showed they could make incredibly flat roads as well. It is likely they can apply the same incline to a canal. So they can't just build it, they can be very precise as well. $\endgroup$
    – Trioxidane
    Commented Nov 18, 2020 at 19:17
  • 2
    $\begingroup$ @JustinThymetheSecond: It wasn't one long continuous trench. It was a complex system of trenches, some 20 km wide on the average. At any given point both the Anglo-French and the Germans had dozens of more or less parallel trenches. The commonly agreed estimate is that the total length of trenches on the western from was about 55,000 km (35,000 miles). $\endgroup$
    – AlexP
    Commented Nov 18, 2020 at 19:18

Our local canal, the Noordzeekanaal was dug and build in the 1800's, the work took about 10 years and only a small part of the 25 km was actually dug by hand.
About 6.5 km (4 mile) through sand dunes was dug by hand and it was where a brook went most of the way through the dunes so not much height to dig. That bit took a lot of effort and about as long as the rest dig/build through a lake, by dredging with whatever methods were available at the time, before steam was actually used in this kind of work.

As you can see in the Wikipedia link (to the canal name above) they dug a lake out deeper, used the soil from the bottom to build dikes to either sides and reclaimed the land to the outside of those dikes to use for fields.

The ground in the area is either sand (dunes,) peat (the land outside the former lake and dunes) and sand or clay at the bottom of the lake.

Once the initial canal was build/dug enlarging was done by dredging and removing the dikes and building new ones.

What you can learn from this is the time it took (building by hand through the dunes) and how to enlarge when you have a basic narrow canal.

Here the builders were lucky enough not to need to build over hills (or build aquaducts to build over valleys.) If you need those you will have to add years to the results.

How many years your canal will take depends mostly on the number of people you can put to work and how much money is invested, using simple machines or animal power will help, how much depends on the tech and animals used as well as the money invested.
It will make more sense to dig such a canal if it connects a series of communities or enlarges a river/brook which is already there but not useful (enough) for shipping. (Rather than to build through an area where there is nothing.)

  • $\begingroup$ Yes, you hit the nail right on the head. The time it takes completely depends on the will to spend the wealth needed to build it, even if the only wealth expenditure is in manpower that thereby can not used for some other economic purpose. The trench on the Wester Front was built using readily available wealth - the manpower was already there, not going anywhere else. The only wealth expendeture was the cost of shovels. $\endgroup$ Commented Nov 19, 2020 at 2:02
  • $\begingroup$ @Justin Thyme the Second What I was actually asking was how much improvement "mechanisation" like cranes and horses would provide compared to the basic man shovel and a wheel barrow. I get the impression that it would not be a lot, but on a large project such as this would, I suspect, have some effect. Obviously the time it takes depends on the size of the work force etc etc. $\endgroup$
    – Slarty
    Commented Nov 21, 2020 at 23:28
  • $\begingroup$ Horses, or some sort of beast of burden, would definitely make the project go a lot faster. Especially when the terrain is flat and unobstructed. A human can move 8 cubic meters of soil in a day. a horse pulling a wagon can move it in a fraction of an hour. So, say one horse equals 32 men, in moving the soil, but I have never seen a horse that can DIG better than a human. As far as I know, humans are the best means of digging until steam power and steam shovels came along (except maybe for hydroblasting). But even then, it took diesel to finally make a pronounced difference. $\endgroup$ Commented Nov 22, 2020 at 1:01
  • $\begingroup$ Plowing, however, is another thing. A beast of burden can plow in soil that a human could never do. $\endgroup$ Commented Nov 22, 2020 at 1:02

You must log in to answer this question.

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