1,135
edits
Luthic: Difference between revisions
m
→Swadesh list
Lëtzelúcia (talk | contribs) m (→Bibliography) |
Lëtzelúcia (talk | contribs) m (→Swadesh list) |
||
| Line 5,194: | Line 5,194: | ||
|- | |- | ||
| '''52.''' valþu [ˈβal.θu] “forest” || '''104.''' þagcare [θɐŋˈka.re] “to think” || '''156.''' staenu [ˈstɛ.nu] “stone” || '''''208.''''' ''Râsdifice'' [ˈraz.di.ɸi.t͡ʃe] “''Linguifex''” | | '''52.''' valþu [ˈβal.θu] “forest” || '''104.''' þagcare [θɐŋˈka.re] “to think” || '''156.''' staenu [ˈstɛ.nu] “stone” || '''''208.''''' ''Râsdifice'' [ˈraz.di.ɸi.t͡ʃe] “''Linguifex''” | ||
|} | |||
Creating word lists depends on the decay of morphemes or changes in vocabulary. For glottochronology to be applicable to a language, the rate of morpheme decay must remain constant. This has led to criticism of the glottochronologic formula, as some linguists contend that the rate of morpheme decay cannot be assumed to be consistent over time. American linguist [[w:Robert Lees (linguist)|Robert Lees]] acquired a value for the “glottochronological constant” ('''r''') of words by analysing the known changes in 13 pairs of languages using the 200-word list by Swadesh. He calculated a value of 0.805 ± 0.0176 with 90% confidence. Swadesh obtained a value of 0.86 for his 100-word list, with the higher value reflecting the exclusion of semantically unstable words. This constant is related to the retention rate of words by the following formula: | |||
[[File:Glottochronologic_constant.png|frameless]] | |||
''L'' is the rate of replacement, ''ln'' represents the [[w:Natural logarithm|natural logarithm]] and ''r'' is the glottochronological constant. | |||
The basic formula of glottochronology in its shortest form is this: | |||
[[File:Divergence time (short).png|frameless]] | |||
''t'' is a given period of time from one stage of the language to another (measured in millennia), ''c'' is the proportion of wordlist items retained at the end of that period and ''L'' is the rate of replacement for that word list. | |||
[[File:Divergence time (long).png|frameless]] | |||
By testing historically verifiable cases in which ''t'' is known by nonlinguistic data (such as the approximate distance from Classical Latin to modern Romance languages), Swadesh arrived at the empirical value of approximately 0.14 for ''L'', which means that the rate of replacement constitutes around 14 words from the 100-wordlist per millennium. This is represented in the table below. | |||
{| class="wikitable" | |||
|+ Glottochronology Time Scale | |||
|- | |||
! Rough Median Dating !! Median Cognate Retention in 100-Word List | |||
|- | |||
| 500 BP || 86% | |||
|- | |||
| 1000 BP || 74% | |||
|- | |||
| 1500 BP || 64% | |||
|- | |||
| 2000 BP || 55% | |||
|- | |||
| 2500 BP || 47% | |||
|- | |||
| 3000 BP || 40% | |||
|- | |||
| 4000 BP || 30% | |||
|- | |||
| 5000 BP || 22% | |||
|- | |||
| 6000 BP || 16% | |||
|- | |||
| 7000 BP || 12% | |||
|- | |||
| 8000 BP || 9% | |||
|- | |||
| 9000 BP || 7% | |||
|- | |||
| 10000 BP || 5% | |||
|} | |} | ||