Quite a lot to reply to, so I'll do the best I can!
@Alex: The only thing Angosey's really lost is pronoun conjugations. It's
gained affixes and gender particles. I think this is because it was a very
stripped down language in 2002, so the only real direction it could go was
up. About the only aspect it had back then was tense (and very simply:
present, past, and future, no marking for progressivity, etc). As I
learned more about linguistics, I filled in my conlang's grammatical
The vocabulary is highly invariant, because Angosey words get "canonized."
Pretty much once I've coined a word, it's in the lexicon for good. Partly
this is because I am more interested in grammar than phonology, especially
at first - it did not matter how it sounded. Another reason is
sentimental. Angosey is a bit of a "heartlang" and acts as a key to
certain past experiences (see my post
http://glossarch.wordpress.com/2013/02/09/backtranslation/). I am loath to
discard words because many of them carry associations beyond their stated
I see what you are driving at, but I'm not sure I agree. One could argue
(as you do) that the very act of using "glossarch" instantiates it as a
word, but I don't think it does, particularly because I present it as a
technical term in the English language. In other words, "glossarch" is not
a word in a conlang - I'm using as if it were actually an English word! If
people on the conlang list started saying they were glossarchs, then I
think we could call it a legitimate word. But a single person imposing a
word on an existing language? I'm not sure.
Having one poem with multiple translations is certainly convenient, but I'm
sure you could go through old texts and extract examples of linguistic
elements appearing and disappearing. I have a very careful analysis of
this in Angosey, and the amount of failed inventions, reversals, and
reinstitutions would make any language reconstructor weep. I think it has
a lot to due with my creative process: I'm a 'language improvisor' rather
than a language creator...the process is necessarily messy.
As for ergativity, noun classes, etc, it's true that Angosey's a bit of a
'kitchen sink' but the sink's been used and abused so much that truly
outrageous constructions tend to fall by the wayside.
Can you cite specific instances for Olaetian phonology? Did you start
studying non-romance languages, and pilfer phonology from them, or did you
formally study phonology? What was the transition process from the early
naive phonology to the more polished inventory it has now?
As for noun genders:
I originally instituted six noun classes to make coining words easier. I
figured one root would instantly have 6 possible words: (physical, emotive,
situational, temporal, locative abstract facets of one concept). Major
This fell on its face, however. What's the locative aspect for a root that
means "eye"? I claimed that would be "watchtower" but I don't think that
necessarily follows. It could be a general word for lookout or view, but
then you need a specific word for "watchtower", etc, etc.
So I get to cheerfully swap noun categories around in a poetic sense, but I
do have to lock in some definitions to keep the language coherent.
2013/2/18 Herman Miller <hmiller@...>
> On 2/16/2013 11:12 PM, Daniel Bowman wrote:
>> Hello All,
>> Last weekend Alex Fink, Herman Miller and I met over brunch and had a
>> great discussion about conlangs. I look forward to bringing up
>> several things we touched on during the conversation. The first is
>> the evolution of my conlang Angosey over the last eleven years. I
>> mentioned that I retranslate the same poem every so often, and I can
>> trace the changes that have occurred (both phonological,
>> orthographical, and grammatical) in Angosey via my past translations.
>> Alex mentioned that he'd be interested in hearing more about it, so
>> I've written a blog post that shows examples from this poem that
>> highlight the evolution of my language. Here's the link in case
>> other list members are curious:
> That's one nice thing about having the same language for 11 years. Since I
> haven't been developing Jarda continuously, I study the old texts for
> examples of "correct" grammar and vocabulary usage. So Jarda hasn't evolved
> in the same way. I can compare different translations from different time
> periods (Relay 3 and Relay 18), but I haven't got examples of retranslating
> the same text. I do have examples in Tirelat of texts that I edited to keep
> up to date with the language as it changed, but a retranslation would be
> more interesting.
> (Relay 3 has an interesting example of a postposition "śa" in Jarda, but
> this could be a result of poetic word order.)
> From the glossarch page:
> The Angosey of Oct0ber 2002 (the first time I translated this poem)
>> had no sounds that English lacked. The word order was different, but
>> I was still stuck in the English/Romance language paradigm. There
>> was nothing truly new about it.
> That pretty much describes Olaetian in the early stages; it had definite
> Romance influences. The only non-English languages I knew much about at the
> time were French and Spanish, so that to me was just how "foreign"
> languages worked. Over time Olaetian acquired all sorts of foreign sounds
> and a few non-Romance features like noun cases, but it fundamentally still
> looks something like a Romance language.
> I was creating all sorts of sketchy languages back then, only a few as
> well developed as Olaetian, but it was nice to have a lot of languages to
> pick from when I discovered a new sound or a new grammatical feature. When
> I found out about ergativity, for example, I created a new ergative
> language, Kazvarad.
> The Angosey of February 2013 does not sound like any version of
>> English I know of. Its phonetic repertoire spans Europe, Africa, and
>> Asia. The grammar is richer and reflects specific ontological
>> choices that suite my way of thinking, from the distinction between
>> emotive and non emotive speaking, to its noun categorization, to its
>> ergative/absolutive verb system.
> We were talking about how ergativity was a conlanging fad for a time, but
> in a number of ways it seems more convenient than the usual
> nominative-accusative system. I mentioned that Jarda had the same word for
> "fall" and "drop" but I must have been thinking of another language, since
> it appears to be a gap in the Jarda vocabulary! I'll need to fix that. But
> what it does have is a single word for "die" and "kill" (rav).
> Angosey has also acquired six noun classes: “au ziramei” means “the
>> (physical object) ‘pearl.’”
> I tried that with Tirelat, but it didn't work out. Maybe it would be a
> good idea for a new Sangari language. Jarda has a set of various
> classifiers (like in Chinese or Japanese), so maybe some other member of
> the Jardic family has reduced these to a fixed set of noun classes.