Re:Where have I heard this before? (Score 1)

The fact that crows can count without having language makes an argument that Sapir-Whorf has nothing to do with this.

The other fact is that the strong version of the Sapir-Whorf hypothesis is dead. If language influences us, it is in a much more subtle way than what we're seeing in this article.

I'd recommend The Number Sense to anyone who is aware that Sapir-Whorf is gone, gone, gone. Dehaene explains how the lower-level numbers (1, 2, 3) are built into our cognitive systems at a most basic level, whereas anything above that is just "many" (he illustrates cross-species differences with examples such as crows counting to seven). His most convincing example outside of experiments is the expression of numbers in various languages.

In Japanese, the kanji for the first 3 numbers are one stroke, two strokes, and three strokes. only on the forth number does it increase.

In Cuneiform, wedge-shaped strokes are expressed in columns and rows with a maximum value of three.

Roman numerals, I, II, and III... then IV.

In current arabic numerals, supposedly, 1 is a single line, 2 is two lines with a connecting stroke, while 3 is 3 lines with two connecting strokes.

This page illustrates the supposed evolution between a few different number sets (Tamil, Hindi, Brahmi, early Arabic) and if you look at the pictures you'll see they all seems to grow out of counting strokes for 1-3, but 4 is a completely different character.

Trinary, anyone?