Log in

No account? Create an account
DeWitt on the Untranslatability of Games - Impressions and Expressions of Ijon — LiveJournal
August 16th, 2009
07:33 pm


Previous Entry Share Next Entry
DeWitt on the Untranslatability of Games
Here's the ever-stimulating Helen DeWitt in an interview:
I'm interested in the fact that natural languages are translatable but games are not. You can translate from German to English, from English to Chinese, and you can use one language as a metalanguage to talk about features of another that it does not share. You can't 'translate' bridge into poker, or poker to chess. Someone who plays bridge well will 'see' the possibilities in a hand in a way that's instantly comprehensible to another bridge player, even if the two players don't share a natural language - but which can only be explained to a non-player by teaching the game.

Our social practices aren't as well developed as our games. A community of game players improves the standard at which a game is played over time. Bridge has only been around for a bit over a century, for instance, but well-developed bidding systems enable even very weak players to communicate the strength and shape of their hand and determine whether they have a good fit with their partner; the systems work well because they have been developed by first-class players who have a good sense of which hands play well. (You can't know the potential strength of a pair of hands, obviously, unless you know what can be done with them.) So if I'm playing bridge and have a six-card heart suit and 3 Aces a King and a Jack I have a very good chance of finding out whether my partner has a) a four-card heart suit and a fistful of honours, b) a four-card heart suit but a weakish hand, c) no hearts, a long spade suit and a fistful of honours, d) a few honours and no strong suit, or e) zilch. (to name just a few possibilities) By way of contrast, we have no comparable sophistication in the communication of sexual preferences or strength of interest. To the best of my knowledge, no one has ever gone to jail for optimistically raising a 1 heart opening to slam on a hand with a singleton heart and the Jack of diamonds; we might think that the sophistication of the game could usefully be transferred to areas of life where the penalties for misunderstanding are higher.

--from an interview by untoward

(3 comments | Leave a comment)

[User Picture]
From:Daniel [oeconomist.com]
Date:August 17th, 2009 07:55 am (UTC)
Well, as an economist, I'm naturally inclined to model what Ms De Witt calls social practices as games. And these are often far more challenging games for which to develop strategies and tactics, which difficulty ought to inform our notions of where strategies are more or less developed.

In fact, bridge is a game that does require translation. Ms De Witt accurately refers to bidding systems, and a player thoroughly familiar with one system but unfamiliar with another may be an ineffectual player when teamed with a player of reversed familiarity.

Also, my girlfriend (the avid bridge player in our life partnership) informs me that a man was once shot repeatedly by his wife for bidding badly (and she even managed to avoid a criminal conviction).
[User Picture]
Date:August 17th, 2009 07:34 pm (UTC)

Between Natural and Game

Games are not translatable because their vocabulary is limited. Game talk can describe in detail relevant scenarios, like one Eskimo language has 400 words for snow. But it was not meant to describe anything else.

Turing complete computer languages can describe each other pretty well, and they are not natural languages. This is because they were made to express a wider selection of scenarios.

Now let us take this farther then intended, as Abbot takes the triangle in Flatland first to lower dimensions, and then to dimensions more than he, and even I, can imagine.

Our own natural language limits us. When we acquire terms in a new field, we limit our mind to analysing using those terms only. May it be a selection of phobias or or physical effects. Most of us will not look for a new physical effect to explain the result they saw, but seek the answer in known effects and their interpretation. We think only what can be expressed in words, otherwise we call it "a feeling". We do not even express all the sounds we could have - a born baby can pronounce every language, but we educate it and limit its sound vocabulary to express only the sounds in its mother tongue.

When we give knowledge, we take away options for new knowledge.
[User Picture]
Date:August 18th, 2009 05:34 am (UTC)

Re: Between Natural and Game

Project Ben-Yehuda [Hebrew] Powered by LiveJournal.com