Glossotechnia playtesting report: simplified deck and experimental new goal and conculture rules

Yesterday I played Glossotechnia with my brother and some friends in
Athens, using the simplified deck (no Phoneme or Syllable onset/rime
or Phonemic Contrast or Suprasegmental cards) and a simple variant of
the experimental new rules I've been discussing offline with Alex
Fink, Sai Emrys and others.  We tried out the conculture rules, where
each player contributes some detail about the culture of the
gamelang's speakers at the end of their turn after coining a new word,
and there were cards in the deck that let you change something about
the culture (normally such contributions have to be additive and
non-contradictory of previous contributions), a First Contact card and
a Revolution card.  Instead of drawing translation challenge cards
with subjects and predicates on them to form translation challenge
sentences, we started out by going round once with each player
contributing a goal for people to try to satisfy with sentences in the
game-language, and the other players deciding how many points it would
be worth.  We came up with these primary and secondary goals:

* conveying the meaning of a new word entirely with one or more
  sentence(s) in the game language, without charades or pictures: 50 pts

* correctly grasping the meaning of such a word from the sentence(s)
  another player uses, 20 pts

* construct a sentence of at least four words with internal rhyme, 100
  pts

* the first time another player uses a word you created in a sentence
  that gets them points, 5 pts per word

* recognizably translate a quotation, 75 pts

* correctly recognize someone else's quote, 45 pts

* create a play on words (spoonerism, pun, etc.) 75 pts

* add a concultural element using the gamelang, 50 pts

* correctly interpret another player's concultural contribution in the
  gamelang, 40 pts

Each of those point-values could only be obtained once for each
player.  Most of the players got points for coining a word using only
the game language; two got points for translating and recognizing a
quotation respectively; one got points for composing a proverb in the
game language.

The game ran between 3.5 and 4 hours, and lasted 33 turns, if I'm
counting correctly.  We tried another experiement where, instead of
keeping one lexicon that gets passed around for different players to
look at, each player maintained their own copy of the lexicon.  This
created an interesting situation, the opposite of Kalusa, where the
players were fairly well agreed on the pronunciation of the language
but each one devised their own orthography for it!  When I played a
sound shift card to merge /U:/ into /u/, a couple of players were
puzzled because they'd been representing both sounds with the same
grapheme, probably "oo".

I was surprised at the way some players used the sound change cards.
In this deck, where there are no Phoneme cards, the Sound Shift cards
have text like this:

    "Specify a sound that occurs in one or more words, another sound
    that will replace it, and optionally a limiting context."

Imprimis, several players used this (and the Insert Sound (epenthesis)
card) to affect whole syllables at once, not individual phonemes, and
not surprisingly the shifts they came up weren't particularly
plausible (e.g. /ma/ shifting to /kEl/ all at once).  Secundus, they
interpreted "limiting context" to mean, not phonological context, but
semantic or grammatical context; so one player used it to coin a
couple of affixes in addition to the word he coined on his turn,
either "epenthetically" creating a syllable ex nihilo in a certain
semantic context or sound-shifting/deleting a phoneme in the basic
verb suffix /ras/ to get /ra/ past tense and /ras:/ future tense.

The syntax of this game-langauge was the most interesting of any I've
seen in a game with non-conlangers; three different Secondary Word
Order cards were played, resulting at once point in this syntax:

VSO for questions
OSV for statements with a feminine human subject
SOV for statements with a masculine human subject
SVO for statements with a nonhuman or inanimate subject

Late someone sent the OSV card to the discard pile and made
SOV used for any human subject.   (In a few hours someone will post
pointing out an anadewism for this unique-as-far-as-I-know syntactic
marking of gender.)

The conculture devised during the game was more interesting than the
game language, not surprisingly; and, what's better, people's
contributions to the conculture were tied into the game-language more
often than I expected.  (Sai, Alex and I wondered if this element
might wind up creating two virtually unrelated games running in
parallel, but it didn't happen, at least not on this occasion with
these players.)  For instance, one player used an "extend meaning"
card to polysemify (polyseminate?) the word for "thief" to mean
"lover" as well, and then told a racy story about why the word has
that double meaning in this culture.  She later explained the loss of
syntactic gender as a social-engineering decree by the priestesses of
the language academy.  Another player described an aspect of the
speakers' religion as involving devotion to a totem animal similar to
an aurochs, and later coined a word for said animal.  I think about 10
out of 33 turns involved a concultural addition tied into the
game-language or a new word or syntax change that closely tied into
the conculture.


Feedback from the playtesters:

Everyone said the syntax cards and sound change cards come up too
often relative to other cards in this deck, and the resulting
too-frequent changes in syntax and lexicon make it harder to follow
what's going on.  One player dropped out several turns before the end,
having become bewiildered by one sound shift too many that, he
suspected, left his lexicon inconsistent with other players'; he said
the frequent sound-shifts made it hard to learn the words already
coined and be ready to use them in sentences.  They suggested several
ways to expand the deck and balance the proportions:

* more Constraint cards (which let you oblige another player to coin a
  word of specified part of speech or in a specified semantic domain
  on their next turn), and more varied: some constraint cards have a
  specific part of speech or semantic domain printed on them, others
  are wildcards and let the player using them specify what kind of
  word the target player has to coin.

* cards that allow you to coin two related words on your turn instead
  of one; maybe some of them are general and some specify the
  relationship between the two words, e.g. antonym, synonym, verb and
  related tool or agent or patient nominalization, etc.

* more Free Pass cards (letting you use English to define a
  word/affix) specialized for classes of morpheme that are hard to
  convey with charades or drawings: grammatical affixes, adpositions,
  conjunctions, maybe abstract nouns/verbs/modifiers.

* cards to let you create a polysemous word on a single turn by doing
  two separate charades or pictures for its two more or less unrelated
  meanings

* more Culture Change cards, e.g. New Technology and War

I mentioned the possibility of adding cards for core words such as the
Natural Semantic Metalanguage word list, and they liked the idea; we
decided it would probably make sense to have about 10 cards with 6
concepts each on them, and let the player choose which concept to coin
a gamelang word for when they play it, rather than double the size of
the deck with 60 individual NSM concept cards.

One player suggested some rule or card mechanics to slow down or
discourage changes of syntax.  We came up with a possible solution:
remove the Secondary Word Order cards from the deck and allow adding a
secondary word order any time you have a syntax card in your hand, but
require use of a rare Change Word Order card to replace one word order
with another.

Alternatively, there could be a second mini-deck with just three cards
marked Subject, Object, and Verb, which are laid out in a random order
to start with.  In the main deck there would be "change word order"
cards to swap the order of any two adjacent constituents.  (Maybe this
mini-deck could have cards for Indirect Object and Locative/Temporal
Complement, etc., as well...?)  But that doesn't allow for secondary
word orders for questions, subordinate clauses, etc., except with a
syntax wildcard perhaps.

All the players liked the free-formedness of word-coining without
reference to a set of phoneme and syllable structure cards in play.
One said, "I feel like our language showed our personality more than
when we had to put sounds together."  Some (all these players had
played at least once before, some two or three times) said it was
the best Glossotechnia game they'd been in yet.

--
Jim Henry
http://www.pobox.com/~jimhenry/conlang/fluency-survey.html
Conlang fluency survey -- there's still time to participate before
I analyze the results and write the article