Last night I went to a ballroom dance lesson, which set me thinking about the Sapir-Whorf hypothesis.

When you are a rank beginner, the natural thing to do is to speak as you are moving: "left-forward-back" and so on. But this of course confuses your partner, who must do the opposite. You can use objective locators - "to the bar", "to the window" - but whilst these work OK for the rumba, they cause a car-crash in the waltz.

What we need, I thought, are observer-neutral spatial terms - something that means "my left/your right."

Although such terms are lacking in English, they do exist in other languages. The Guugu Yithimirr of northern Queensland use points of the compass rather than subjective terms such as left or forward; they would say, for example, "the tree is west of the house."

I suspect that our ability to learn dancing would, in the initial phase at least, be enhanced if we could use such language and the thought that accompanies it - the ability to identify compass points.

This is where the Sapir-Whorf hypothesis enters. It is the theory that our language constrains our thoughts - that, as Wittgenstein said, "The limits of my language mean the limits of my world." For now at least, the English language is constraining my ability to learn to dance.

Maybe the limits on our thought imposed by subjective terms such as "left" and "right" are tighter than this. Here's a conjecture. Perhaps being accustomed to think of space in subjective language creates or intensifies an individualistic, egocentric bias in our general thinking - a tendency to think that our point of view is the objectively correct one. The fact that there are cross-cultural differences in degrees of overconfidence, perhaps because of differences (pdf) in "cognitive customs" is compatible with this hypothesis.

Here's another example. In English, we speak of the future as being in front of us; we say, for example, "I'm looking forward to my holiday". This encourages us to think that we can see the future; we can, after all see what's in front of us. But this is not the case. In important respects - such as recessions - the future is unknown and perhaps unknowable.

Here, the Aymara people have the advantage on us. They speak of the past as being in front of them and the future behind them. This makes sense: we can see the past better than the future.

Take another example. We English have no precise equivalents of the ancient Greek words for "phronesis" or "arete". It's possible that this lack is related to Alasdair MacIntyre's complaint that our moral thinking is much more confused than that of the Greeks.

Perhaps this point generalizes. It's not only language that constrains our thought but our conceptual schema. One reason why culture wars are so bitter in religion or even in macroeconomics is that different cultures have different standards of what matters - and the more bone-headed partisans don't appreciate this.

This is not to say that the constraints imposed by language and culture are tightly binding; translations are after all possible and we can construct some meaning of words like "arete". What I'm saying, though, is that we often fail to appreciate these constraints for the same reason that fish don't know that they are wet.

I guess that all I'm calling for here is a little less egocentricity in our thought and a little more effort in trying to understand where others are coming from - because perhaps our minds aren't as open as we think.