You sometimes hear people say that syntax is not important. Not true:
syntax is one of the most important things there is. (Maybe what they really
mean is that *their* syntax is better than yours...) They probably
refer to the fact that there are multiple ways to write down the same model.
That is true, but it by no means follows that all ways are equivalent. It is a
philosophical debate whether a person's language determines his mental model
of the world or whether the language is merely the surface structure of a
deeper model that is essentially the same for everybody. For human languages
we leave the debate to others, but for our formal, limited languages we must
follow the advice from mathematics: there is nothing like a suggestive
notation for understanding the model.

The standard example from mathematics is d`t`/d`x`, which
looks like (and is treated by physicists as) an infinitesimal change of
`t` divided by an infinitesimal change of `x.`
Mathematicians can tell you exactly in what ways it behaves like a division
and it what ways it doesn't, but the fact is that for nearly all formulas that
occur in physics it *does* behave like a division. And it makes life
very easy for physicist that they can simplify
(d`x`/d`t`)(d`t`/d`y`) to
d`x`/d`y`. It is easy because it *looks* the same as
simplifying (2/3)(3/7) to 2/7. The notation for differential equations hasn't
always been like that, but it is not for nothing that all other notations are
forgotten.

Learnability, then, has a lot to do with readability.

- [Wittgenstein86]
- Wittgenstein, Ludwig. Tractatus logico-philosophicus. Amsterdam. Dec 1986. Athenaeum-Polak & Van Gennep. In German, with Dutch text on right-hand pages, translated by W.F. Hermans