On Mon, 19 Jan 2015 12:50:22 -0800, Gary Shannon <[log in to unmask]> wrote:

>Thinking back on Terry Winograd's "Blocks World" for his research conlang
>SHRDLU ( ) it occurred to me that I
>often hit a point in my conlang building when I start feeling overwhelmed
>by the sheer volume of lexicon that seems to be needed to do any
>significant translations. 

From here, with the computery example of SHRDLU on the brain, I thought I'd predicted where you were going with this: write a conlang to be spoken in the world of a computer game!  Nearly any game will have a strictly limited set of objects and characters in it, and a strictly limited set of ways that they can interact and you can interact with them; there's the bounds of your lexicon right there.

>Then another odd idea occurred to me. Assume this microcosm has nothing
>that functions as a "market" and that the economy is a gift/debt economy
>that works as follows: [...]

Mm, I came up with negative-valued money in elementary school, but it didn't take long to identify the flaw: if you have some antimoney ($b ?) the economically rational thing to do is to lose it like a hot potato.  Throw it in the sea or something.  

In a Dunbar-sized society you could avoid this beign a problem by remembering who's supposed to have how much antimoney.  But having done this, you quickly degrade to the state of not bothering with the money at all, and just keeping track of who owes the community how much.  There's good precedent: the communalisation I can't speak to, but that aside, treating debt as only a social obligation with no physical markers is extremely common if not universal in premercantile, pretaxation, etc. natcultures.  (Adam Smith, positing instead that the stage before money was barter, object for object only, has empirically been shown quite categorically wrong.)  Certainly, if I was paring down a culture for whatever reason myself, I'd abandon money long before doing anything so incomprehensibly utopian as the banishment of all disease.