Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Map and territory: Natural structures

1 Post author: casebash 01 August 2017 01:53PM

This will be a very short post which simply defines one term which I find useful when discussing the map and the territory.

I find it very useful to have a term that helps clarify that the map is not completely arbitrary and that there are things in the territory that are natural candidates for appearing in the map. For example, for the Ship of Thesus, one natural candidate is the pure, original, unmodified ship; another are the fixed percentages (ie. 50% original); another would be a continuity based measure. If you are asked to create a definition of what counts as the Ship of Thesus, these are some of the first ideas that you would come up with, although you would of course need to define it in much, much more detail to get all the way down to the level of the territory.

Or suppose you are trying to define what is meant by table. Again, the definition is purely arbitrary and whatever you choose, but there are certain natural structures in reality that pop out at you. One might be all four-legged, non-living objects with a flat top, another might relax the four-legged requirement so that it only required four legs at one particular time, ect.

When I'm explaining that a particular concept has been reified, it greatly clarifies my position to explain that I don't believe that the concept is empty, but there is *something* behind it that leads us to want that word. That something is really not a single thing (or else it would be real, not reified), but a collection of closely related 'natural structures'. Each of the definitions provided for the Ship of Thesus or a table corresponds to a different natural structure, while the term itself appears in the map. I hope you find this word useful too, but if you have any suggestions for a better term, please mention it in the comments.

Comments (9)

Comment author: Dagon 01 August 2017 05:54:30PM *  3 points [-]

I agree that there's something which draws us toward similar map structures. However, I'd avoid the word "natural" for describing this, as I suspect it's cultural or shared-experience rather than inherent in the territory. I therefore suspect that how one identifies "us" will result in very dissimilar map choices. Thee's likely to be much divergence between the categorization by ML models vs humans, and some variance between humans with wildly different cultural norms.

I might say "cultural shared expectations" as the thing you're trying to name.
edit: or maybe "comfortable map structures" or "focal structures"/"Schelling structures" (cf. https://en.wikipedia.org/wiki/Focal_point_(game_theory) ).

summary: I think these structures are from similarity of mapmakers, rather than anything in the territory.

Comment author: ChristianKl 03 August 2017 01:30:15PM 1 point [-]

Your concept of a table is not the one that's used by most people. Most people don't have a problem imagining a 5-legged table and call an item like http://www.ikea.com/us/en/catalog/products/10135659/ a stool instead of a table.

Comment author: casebash 03 August 2017 02:00:52PM 0 points [-]

It's just an example.

Comment author: ChristianKl 03 August 2017 03:44:17PM 0 points [-]

I think the default way people tell a stool from a table is that a stool is something on which you sit and a table is something on which you don't sit but put stuff.

It's not about whether the surface is flat or the amount of legs. You seem to argue that the identity should be defined about such "natural" or objective qualities without really making a case for it.

Comment author: Elo 03 August 2017 08:16:12PM 0 points [-]
Comment author: ChristianKl 03 August 2017 01:27:28PM 0 points [-]

Things in the territory don't appear in the map. They have a representation in the map.

Comment author: casebash 03 August 2017 02:00:21PM 2 points [-]

Yes, they don't appear in the map, but when you see a mountain you think, "Hmm... this really needs to go in the map"

Comment author: Dagon 04 August 2017 04:32:39PM 0 points [-]

There are LOTS of maps which don't include mountains.

Comment author: ChristianKl 04 August 2017 08:50:13AM 0 points [-]

"Go in" is something different than "represented by". It's worthwhile to be conscious of the abstraction.