One of last week’s Google doodles was in honor of the 115th anniversary of the discovery of the Antikythera mechanism. This is a fascinating device that many people consider to be one of the earliest analog computers (using physical processes or quantities instead of symbols to keep track of information) and it definitely deserves its own post, but today I’ll focus on something I found in my Antikythera research. This list of ancient computing hardware includes a few prehistoric artifacts, the abacus (c. 2400 BC, thanks to the Babylonians), and next on the list, the Chinese south-pointing chariot.
To clarify, this isn’t a chariot that always faces south, which would be great for one-way trips but would create a certain demand for north-facing chariots. The south-pointing chariot (or carriage, in some sources) moves in any direction but has a small figure mounted on the top. This figurine incorporates a mechanism that points south regardless of which way the chariot itself is facing.
That sounds pretty basic to us, even if it’s been a while since you used a compass yourself instead of outsourcing the work to your phone. Consider, though, that magnetic materials weren’t used as compasses until around 1000 AD, also in China. The south-pointing chariot, according to historical references, may already have been around in some form by about 1000 BC, and we have reliable sources describing one created by Ma Jun around 200 AD.
Without a magnetic compass, how did the south-pointing chariot work? There are a few theories, which may all be correct, as different designs almost certainly have existed over the centuries.
The most popular theory is that the direction-finder used differential gears. This is a system used in most cars today that involves three shafts connected by gears, allowing the wheels on opposite sides of the car to rotate at different speeds, as is necessary when making a turn, while being driven by the same engine.
In cars the engine shaft rotates at a speed proportional to the sum of the wheel rotation speeds, but for the south-pointing chariot, the system would work slightly differently. The third shaft, rather than going to an engine, would lead to the south-pointing figure. Instead of the rotation speed being related to the sum of the wheel rotation speeds, it would be related to the difference—if the difference is zero, the chariot is going straight, and the figure won’t rotate at all. Any difference means some degree of turn, and the gears would be designed so that the figure-shaft would swivel to compensate for the turn.
If the south-pointing chariot used differential gears, it would possibly beat the Antikythera mechanism as the earliest known use of the idea, depending on when the chariot was actually developed. This is a fascinating possibility, but it’s unproven, since no ancient chariots survive and we’re going off sometimes vague descriptions.
It’s likely that at least some chariots had a different design, also using gears but without the differential element. Instead, pulleys would lower small gears that would link one wheel or the other to the figure-shaft, depending on which way the chariot was turning. This required a fixed turn radius, so there may have been a third wheel to set that radius, but it’s not mentioned explicitly. Still, the design would’ve been simpler than the differential gears.
Whether the chariot used differential gears or not, it would have fallen victim to the inevitable gap between design and construction. You can read a more detailed discussion here, but in short, the width of the wheels would have to be identical to within one part in a million to make the chariot worth using for any journey longer than a few kilometers. The inaccuracy of the pointer would increase with distance and only get worse as wheels wore down in non-uniform ways, as they tend to do when you drive on roads.
With these limitations in mind, many scholars believe the south-pointing chariot was used primarily as a curiosity, to impress or entertain rather than to provide a necessary service. Regardless of how it worked or why it was used, the south-pointing chariot is an impressive piece of engineering, and it’s been reinvented in various forms several times as a testament to the interest it holds for engineers.
Personally, I think it’s especially cool for a few reasons (one being that I’m basically interested in everything, but that’s a personal problem). On the list of ancient computing artifacts, it jumped out at me as one of the few that wasn’t explicitly astronomical. I love astronomy as much as (or possibly more than) the next person, but it’s always fun to see something that bucks the trend.
Also, I love comparing the technology of ancient cultures and seeing what we know about how the interplay of human cultures has spread innovations and ideas across the world. With a lot of early devices like the chariot and the Antikythera mechanism, it’s still unknown exactly how much cross-cultural influence was a factor, but it’s fascinating to try to understand the specific needs and pressures that guided early engineering along different paths.
You can read more about the south-pointing chariot/carriage on many sites, a few of which I’ve linked to in this post. To find out more about differential devices, check out the Wikipedia explanation, this How Stuff Works introduction, LearnEngineering, or this great 1930s tutorial video!