Another one of those big controversies in the cognitive sciences, apparently.

I just read a series of conflicting studies trying to determine the extent to which our native language influences other aspects of how we think about the world, particularly spatial relationships.

Apparently some languages don't have relative language descriptors for "tabletop" space. If you have a pencil on the desk, and it's next to your coffee cup, and I asked you where the pencil was, you would probably say "To the left of the coffee cup" or "Next to the coffee cup". Speakers of some languages in some parts of the world would say "The pencil is east of the coffee cup", using an "absolute" frame of reference (well, that's what it's called in the literature anyway...there is no absolute frame of reference, so it's actually earth-centric, right?).

The hypothesis goes that this attribute of the language will have effects on other aspects of spatial cognition. So in one paper I read, they did a test with little plastic animals. They'd line them up facing a particular direction, say north, with the subject facing them. They'd give the subject time to memorize the order and direction of the animals, then they'd scoop them up and walk the subject to another table, so that the person had turned 180 degrees. Then they'd hand them the animals and say "Make it just like before".

So the idea was, people like you and I, who have a strong preference for relative descriptions of tabletop space, would orient the animals relative to ourselves, so that the nose of the first animal would still be pointing to our left. An Aborigine would orient the animals according to the absolute frame of reference, so that the animals' noses would all be pointing north.

So they said they found this effect. However, another group of researchers said the experiments were flawed...that the tests of relative spatial speakers were done indoors, with no landmarks (like buildings or the sun), while the tests of absolute spatial speakers were all done outdoors, thus biasing the results. They tested a single language group (English speakers) indoors and outdoors and got a difference in effect. The idea being that if you're indoors you're more likely to use a relative frame of reference (yourself as a landmark), and if you're outdoors you're more likely to use an absolute frame of reference. Sounded sensible to me.

The original group went back and tried to replicate the critics' experiments, but were unable to duplicate the results. So either somebody's methodology sucks, or there's some unaccounted-for variable somewhere screwing things up.

I don't have any problem believing that our language influences other aspects of cognition, although on the spectrum of just how much it influences other areas of thought, I'd be on the more conservative end. Spatial reasoning is very, very old (I just read a paper about spiders who hunt other spiders and do quite a bit of spatial analysis and planning beforehand). Language isn't very old, evolutionarily speaking. I would guess it has had considerable impact on memory, especially how we remember facts (as opposed to episodic memory or events). It's a symbolic representation system and thus is a probably a much more efficient way to store and retrieve categories and invididual static instances of things.

But I'm less sold on the idea that it has a major impact on how we orient ourselves or think about objects in terms of space.

Could be, though...research so far is inconclusive.