Joined: 30 Oct 2006
[META][REFERENCE] A brief analysis of the Lanti language
Given what we've seen thus far
As a complement to the Lexion thread, I thought I'd post some of my more general analysis of the Lanti language as a whole. I'm certainly no proffessional linguist, but it IS something I've studied on an enthusiast level.
By my count, we've seen about 200 words of Lanti so far, about 125 of which have been unique. Of those 125, over 75% have only appeared once, which makes guessing on translation by context pretty thouroughly impossible. The translations we've been given either through 746865656e64 or our contacts have elucidated about 30% of the words we've seen, and educated guessing can potentially bring us up to about 40% comprehension. Not great, but not terrible either, given the limited body of text we have to work with.
So, not that statistics aren't fun, but what about the Lanti language itself? What do we know about it linguistically? Well, first, Lanti has essentially the same grammar and syntax as English with an unheard of, nearly 1 to 1 symantic correspondance (OOG: basically, a word level cypher of English). Furthermore, Lanti appears to have little or no internally consistant symantic relations. This is evidanced by the fact that words with closely related symantics (such as "we", "us" and "our") have little phonetic realtion in Lanti, nor is there any particular noticiable relation between similar parts of speech.
OOG: It seems likely (to me at least) that Lanti may be based on some predicably random (i.e. Hash-function like) encyphering of english words. I'm basing this guess on the relative lack of symantic cohesion in Lanti itself, and by the somewhat awkward phonetics of many of the words. (words like "thfr", "kn", "hj", "jkl" etc.) What this means for us, is that Lanti has nothing resembling any linguistic inflection that we can use to guess at the meaning of new words. We're left, unfortunately, with pure context.
"D'uh", right? So what? Well, one bright point that seems to have resulted from this, is that Lanti appears not to distinguish between singulars and plurals. Given Lanti's lack of internal semantic consistancy (OOG: i.e. assuming it to be a randomly generated encyphering), this makes sense, as otherwise words like "man" and "men" could potentially be unrecognizably different.
As evidence of this facet of Lanti, consider "aladi" which was translated in the first recording as both "has" (singluar of auxilliary verb, "to have") and "have" (plural of the same). This is also evident in the word "mrkzai", translated in the first recording as "civilization" (singular) but used on Yshk Lanti before a list of what are presumably multiple civilizations (plural). Finally, assuming we accept "vyl" as the past of "to be", it is also used in both singular ("was") and plural ("were") contexts.
So, to summarize:
- We can understand or make a good guess at 40% of what we've seen
- The other 60% is mostly single-instance words that are restrictively hard to guess by context.
- OOC: Lanti is probably a random(ish) cypher, with no linguistic morphemes or structures to exploit for translation
- Lanti does not distinguish singulars from plurals
I'm afraid that's all I've been able to glean from my (admittadly short) study of Lanti. If anyone else has anything to add, or wants to question / criticize my conclusions, have at it!
Joined: 23 Oct 2006
Lanti word frequency
For completeness here are the current word frequencies. I included the currently known translations down to frequency 3.
15 ny (the)
8 di (of)
6 ra (to)
5 hasya (?)
4 vyl (is/am/were/was)
4 un (I)
4 teyailn (there)
4 srk (you)
4 kol (human)
4 en (and)
3 yshk (for)
3 peynin (?)
3 p (a)
3 kon (?)
3 aladi (has/have)
based on this compilation of all texts known so far:
Ra nnra koldaaj, jal hasya adjnak ny nukvad.
Ra nednai ny takma, peynin hasya keilyn ny moquin. Teyailn vyl esun mrkzai di peynin di Gai, teyailn vyl Lanti, teyailn vyl La, en masi ust awlyn.
Lanti vyl p shantilan lafvyl hj ny tilnlyl ksi acktrol etshundan. Ruk peynin srk esutl smilan thyr jalun srk ra ny ishka di ny eutshoo. Ny atunama tei hyk, yshk srk hasya tollen en trikara kol di kon kanjal katutel. Ny gharsharn kol tolkna monlj j kon fejatala. Mshn klyk shorm kol yshk ny kayk di kon ponkhn.
spiala verkandon un esetula ra actino srk laku teyailn aladi iraja p arpka flalo talm craah olj fhar ra guakala ny linguala di shundaj mrkzai, ghyra aladi monlj hykyar en gora zinhaj sha skorjin ny trakan.
Un ytar un kiyu kynpen ny jzan nympar ikwk esolot alad kn ytar utahyr"
Gilnor osen thfr ij licij tednot lankal beche rawet ivektol ny chrel di p
sonlal nerken. Chisheck gilnor osen tikov ny ishka di ny eutshoo. Jal tei
rean enedia kol en hasya ske tolkna. Craah lysheen un hasya nlyuk yshk
iance. Chisheck shktol aladi masuka goro rnal kn geifyl ra shundaj shalan
Jkl kintala gry ocktel.