In defense of complicated measurement systems

Archaic measurement systems were much more complicated than the systems we use today. Even modern imperial units, while more complicated than metric (SI) units, are simple compared to say medieval English units. Why didn’t someone think of something like the metric system a thousand years ago? I imagine they did. I don’t think they were missing a key idea; I think they had different criteria.

Virtuvian man: human scale measurement

Convenient units

We look back and wonder why people would have, for example, unrelated units of length for measuring the lengths of different things. (Of course these units could be related, but they’re not simply related, not like multiplying by powers of 10.) You might ask a medieval peasant, for example, why cloth is measured in different units than the distance between villages. The peasant might look at you like you’re stupid. “Why would you want the same units for such different things? Are you wanting to buy enough cloth to line the road to the next village?!”

When you buy cloth, you naturally measure it with your arms, and so it makes sense that cloth would be sold in units related to arms lengths [1]. If you’re measuring a long distance, you naturally measure it in units related to walking, or maybe in terms of the distance to the horizon. It was more important that each measurement be convenient for its purpose than for the units to be simply related to each other.

An acre was the amount of land a man could plow in one day. It was almost a unit of work rather than a unit of area. It was a very practical way to measure the size of a farm, and if it had a complicated relationship to, say, the unit of measurement for buying cloth, so be it.

Because the acre was based on the effort required to plow land, an acre was bigger in some regions than others [2]. We immediately think this is a horrible situation. But what did a medieval Scottish farmer care that an acre was a smaller unit in Scotland than in Ireland? Eventually this did get to be a problem, and the acre was standardized. Trade on a wider scale made standardization more important.

Why all these multiples of 2 or 3 between units rather than powers of 10? When you’re doing math in your head, or on paper with Roman numerals, small integer factors are nice to have.

History and compromise

Some traditional units must have seemed unnecessarily complicated even to contemporaries. These may have been the result of history. Two cultures come into contact and have to reconcile their units of measure. Or something about the world changes, such as wider trade mentioned above, making it necessary to compromise between what is familiar and what would be better going forward.

The metric system came out of the French Revolution. The revolutionaries weren’t concerned with history; they were prepared to blow up the world and start over. That didn’t work out well in most areas, but it did work out well for units of measure. The metric system (technically its successor SI) is now used around the world. [3]


[1] See the ell.

[2] Episode 115 of The History of English Podcast has a good explanation of this and many other related topics.

[3] It’s commonly said that the US does not use SI. It would be more accurate to say that the US does not¬†exclusively use SI. The uses of imperial units are obvious, such as highway speeds posted in miles per hour, but SI is used quite a bit behind the scenes.

If you’d like to get daily tweets about units of measurement, follow @UnitFact on Twitter.


8 thoughts on “In defense of complicated measurement systems

  1. People were more focused in ancient time – tailor was not interested in plowing an acre of land, while peasant was not interested in cutting fabrics. They had their trade associations (Guilds?) without “looking over the fence”. It was important that one is understood inside the guild. Still is the same with gold and precious stones – one carat being equal to equivalent to 200 milligrams. One can buy 1000 carats of diamonds, but can not buy 1000 carats of bread.



  2. To be even a bit more pedantic, the US doesn’t use the imperial system – it uses the American system of units. Some units are the same as the imperial equivalents, while others, like the gallon, are substantially different.

  3. I’m curious as to what parts of the French revolution “didn’t work out so well”. It seems to me that the metric system’s adoption was not so much a result of its inherent qualities, but rather a side-effect of the post-revolution French influence in 19th century science and politics.

  4. Interesting reading, John.
    It makes sense that intuition at times might be more relevant than mathematical purity.
    While different lengths of feet seem sorted out, historical left-overs at times still confuse us today. A good example is the difference between US and British gallons (
    At times, the confusion comes at a steep price, for example when losing an orbiter (

  5. As a Brit who lived in the US I like to ask people what you can buy in an American grocery store in metric unit only (IEe not the things that only have metric unit for sales in Canada). Light bulbs (sold in watts rather than horse power) and reading glasses (diopters) are the easiest to think of.

  6. Deki,

    We still use a guild like concept. When you computerize a domain-specific process, that process is surrounded by a conceptual model embedded in what I used to call a functional culture. The real term is epistemic culture. The metrics and thought processes stay within that domain’s epistemic culture. Corporate cultures select one domain among many. They put that domain’s epistemic culture at the center of their decision making. Corporate cultures are not about Thursday being Beer Day. Even firms doing engineering decide that financialization is more important than engineering. Those are cultural fights first.

Comments are closed.