Hi Aurelio,
Here I present a pattern of sound correspondence (PSC) between Old Norse (ON) and Taiwanese (Tw) to illustrate the simplification of word-final consonant cluster. The ON words in this PSC have a word-final -ngr whereas their Tw corresponding words with the same meaning have just an -n ending. The word-final -gr may have been a consonantal syllable in ON. If that is the case, this PSC may illustrate Relationship (2) instead.
In the following examples, the Tw word that lacks a Chinese cognate is designated as Group-1 word (G1W). Those that have Chinese cognates are given the hàn-jī.
PSC-1: ON -angr : Tw -an
rangr ‘wrong’ : loān 亂 ‘wrong’ as in loān-phòann 亂判 ‘to judge wrongly’
angr ‘bay, firth’ : oan 灣 ‘bay, gulf’
angr ‘sorrow’ : oàn 怨 ‘grief, sorrow’
sam-gangr ‘sexual intercourse’ : sann-kàn, sio-kàn 相姦 ‘sexual intercourse’
svangr ‘slim, slender, thin’ : sán (G1W) ‘slim, slender, thin’
vangr ‘field’ : pán 坂 ‘field by the hill or river’
This PSC shows the correspondence of -angr > -an where the vowel -a- remains unchanged. The only change is the simplification of the word-final consonant cluster. However, the first 3 examples have an additional -o- inserted before the -a- in Tw words. The additional -o- functions as a glide (kài im 介音) and is pronounced like the consonant [w]. It is written with -o- according to the orthography of the POJ script that was developed more than 150 years ago based on the Amoy dialect of Hokkien. Thus ‘Taiwan’ is spelled Tâi-oân according to this script. Some other explanatory notes are given below.
Example 1: Tw lacks the [r] sound, therefore, the ON /r/ is substituted with /l/ in Tw.
Examples 2 & 3: Note that in ON, these two words are homophones. Their corresponding words in Tw are also homophones except for the difference in tone.
Example 4: This is another case where both 2 syllables are retained in Tw. This is actually a compound word using the element sam- (together). There are many other examples of sam- compound words in ON that find Tw correspondences, three of which are listed below:
ON sam-jafn (equal to) : Tw sann-siâng 相像
ON sam-fagn (I rejoice with another) : Tw siang-hoan 相歡
ON sam-tal (interview, colloquy) : Tw siong-tâm 相談 (> 商談, 商討)
In the last example above, ON tal (talk) : Tw tâm 談 , there is a sound correspondence between -l and -nasal sound. The word 相談 is also found in Japanese sōdan (conversation, consultation), whereas Swedish samtal means ‘dialog’. Tw 相 has several variant pronunciations: sann, siang, siong and sio.
Also, note that ON g- (as in gangr) often corresponds to Tw k-. Examples:
ON gan (frenzy, frantic gestures) : Tw kán, kóann 趕
ON garðr (house, dwelling) : Tw ka 家 (with loss of the word-final consonant cluster -rðr)
ON gunnr (war, battle) : Tw kun 軍 .
There is something very significant that needs to be highlighted. We all know that sexual intercourse is one of life’s most important functions, as Confucius says, “食色性也 .” Thus, the term for such activities should be in the vocabulary for the core family life, and passed on from generation to generation unless altered by external forces. The fact that the term for sexual intercourse fits the pattern for sound correspondence between ON and Tw suggests that there might be some unspoken, yet-to-be-revealed relationship between the two languages.
Example 5: ON svangr has a double consonant in the word-initial. In Tw phonology consonant clusters are not allowed; they get simplified: sv- > s-. Thus, we find Tw sán (G1W, slim, slender, thin) to correspond to ON svangr. Other examples of the sv- to s- correspondence are:
ON sveinn (boy, servant, squire, page) : Tw seng 生 as in tông-seng 僮生 (squire, page)
ON svinnr (swift) : Tw sìn 迅
ON sver (I swear) : Tw sè 誓
(In my survey of the correspdonences involving s-containing consonant clusters, I found that sometimes s- is retained, sometimes the consonant after s- is retained, yet sometimes both alternatives exist producing 2 “daughter” words. I have not been able to figure out what factors influenced the retention.)
Example 6: v- in foreign words is often rendered in Tw as p-/ph-. A classic example is the Buddhist term nirvana which is transliterated as 涅槃 liap-pôan where -vana is rendered as pôan. So, ON vángr (field) : Tw pán 坂 (field by the hill or river). Other examples of v- to p-:
ON velt (I roll, tumble over) : Tw péng (G1W) as in péng-pêng (turn over) with -l corresponding to -nasal sound (as mentioned above)
ON ván (hope, expectation) : Tw phàn 盼 as in phàn-bōng 盼望
ON vinr (friend) : Tw pêng 朋 as in pêng-iú 朋友 .
All ON words mentioned above are listed in Geir T. Zoëga (1910), A Concise Dictionary of Old Icelandic, Clarendon Press, Oxford. Old Icelandic is generally considered representative of Old Norse. The Oxford edition is no longer in print, but fortunately for us students of Germanics, Dover Publications has recently published a paper-back reprint edition. The dictionary can also be assessed through the Internet from the following site:
http://www.northvegr.org/zoega/
I hope that with PSC-1 illustrating the simplification of word-final consonant clusters, we have taken the first steps toward discerning hidden patterns of sound correspondences between ON and Tw and toward a better understanding of the origins of Tw.
Thank you for your stimulating questions.
Heruler
PSC-1: Old Norse -angr : Taiwanese -an
PSC-1: Old Norse -angr : Taiwanese -an
Last edited by Heruler on Wed Aug 30, 2006 1:18 pm, edited 2 times in total.
An epilogue - Testing of PSC-1 -angr > -an
I am trained as a scientific researcher in a biological field, and so my habits are to collect data first, and while gathering data, if I start to recognize a pattern emerging, I form a hypothesis about the pattern of sound correspondence. With this working hypothesis I try to gather more data, the more the better so that the larger the sample size, the smaller the errors due to random coincidences. This is followed by testing the working hypothesis. A correct hypothesis should be able to make predictions which can stand the test. If the hypothesis proves wrong, I then look for other possiblilities. If the hypothesis proves right, then I continue pursuing the track, and look for related patterns. The crucial part is testing whether the prediction made by the hypothesis proves right.
Let me use PSC-1 -angr > -an as an example to show the process of testing. There is a German word schwanger ‘pregnant’, which has been derived from West Germanic *swangra-, which originally meant ‘schwer, schwerfällig’ (Kluge Etymologisches Wörterbuch der deutschen Sprache, 23rd edition, Walter de Gruyter, Berlin, 1999. Page 747), translated into Taiwanese, tîm-tāng-ê, tî-tūn-ê 沉重的, 遲鈍的. Since it has the -angr ending (ignoring the second syllable because Taiwanese phonology usually takes just one syllable), PSC-1 predicts that, if Taiwanese has a corresponding word relating to pregnancy or childbirth, it should sound something like san (ignoring the tone for now). For this I followed the same rule for simplifying the consonant cluster sw- to s- as that for ON svangr to Tw sán. Does Tw have such a word relating to pregnancy or childbirth? Yes, there is the Tw word sán 產 as in seng-sán 生產 ‘childbirth’ whose meaning is slightly off its German counterpart, but still in the same category. Furthermore, since it is also used in liû-sán 流產 ‘miscarriage’ (= loss of pregnancy), as well as sán-hū 產婦 ‘prenant woman about to give birth’, so we can say that German schwanger : Tw sán 產, but the latter covers a wider semantic sense than its German counterpart. It was truly gratifying to see that the PSC was able to stand the test.
By the way, the word sán 產 as in seng-sán 生產 is listed in Carstair Douglas’ dictionary (1873, p. 409) as well as Barclay’s Supplement (p. 190). Japanese also uses this word san 產. The scientific term for 'obstetrics' hū-sán kho 婦產科 was first coined by the Japanese during the Meiji era.
Incidental notes:
(1) The Old Norse word frú (lady; German Frau) is corresponded by two Tw words hū 婦 and lú 女, due to "alternative splicing" of the word-initial consonant cluster of ON fr- to Tw (f->) h- or (r->) l- so that it forms Tw hū 婦 or lú 女. Through pleonastic formation (pairing of synonyms) of the two "daughter words", ON frú corresponds to Tw hū-lú 婦女.
(2) The Tw colloquial word for "pregnant" is pēnn-kiánn with pēnn being the nasalized form of pēng. The word pēng can be correlated with the first syllable praegn- of the Latin word praegnant (pregnant). Due to Tw phonology, the word-initial pr- is simplified to p-, and the word-stem final -gn is usally corresponded by Tw -ng, for example: ON hegn (I punish) : Tw hêng 刑 (punish). The Latin diphthong -ae- has been known to undergo a sound change to -e- in Romance languages. For example, L. praestare (to lend) becomes Italian prestare, French prêter, and Spanish, Portugese prestar.
Of all the words I have been able to find correspondents in European lexicons, about 50% are related to ON and other Germanics, and the other 50% to Latin-Greek. But that is another story.
Heruler
I am trained as a scientific researcher in a biological field, and so my habits are to collect data first, and while gathering data, if I start to recognize a pattern emerging, I form a hypothesis about the pattern of sound correspondence. With this working hypothesis I try to gather more data, the more the better so that the larger the sample size, the smaller the errors due to random coincidences. This is followed by testing the working hypothesis. A correct hypothesis should be able to make predictions which can stand the test. If the hypothesis proves wrong, I then look for other possiblilities. If the hypothesis proves right, then I continue pursuing the track, and look for related patterns. The crucial part is testing whether the prediction made by the hypothesis proves right.
Let me use PSC-1 -angr > -an as an example to show the process of testing. There is a German word schwanger ‘pregnant’, which has been derived from West Germanic *swangra-, which originally meant ‘schwer, schwerfällig’ (Kluge Etymologisches Wörterbuch der deutschen Sprache, 23rd edition, Walter de Gruyter, Berlin, 1999. Page 747), translated into Taiwanese, tîm-tāng-ê, tî-tūn-ê 沉重的, 遲鈍的. Since it has the -angr ending (ignoring the second syllable because Taiwanese phonology usually takes just one syllable), PSC-1 predicts that, if Taiwanese has a corresponding word relating to pregnancy or childbirth, it should sound something like san (ignoring the tone for now). For this I followed the same rule for simplifying the consonant cluster sw- to s- as that for ON svangr to Tw sán. Does Tw have such a word relating to pregnancy or childbirth? Yes, there is the Tw word sán 產 as in seng-sán 生產 ‘childbirth’ whose meaning is slightly off its German counterpart, but still in the same category. Furthermore, since it is also used in liû-sán 流產 ‘miscarriage’ (= loss of pregnancy), as well as sán-hū 產婦 ‘prenant woman about to give birth’, so we can say that German schwanger : Tw sán 產, but the latter covers a wider semantic sense than its German counterpart. It was truly gratifying to see that the PSC was able to stand the test.
By the way, the word sán 產 as in seng-sán 生產 is listed in Carstair Douglas’ dictionary (1873, p. 409) as well as Barclay’s Supplement (p. 190). Japanese also uses this word san 產. The scientific term for 'obstetrics' hū-sán kho 婦產科 was first coined by the Japanese during the Meiji era.
Incidental notes:
(1) The Old Norse word frú (lady; German Frau) is corresponded by two Tw words hū 婦 and lú 女, due to "alternative splicing" of the word-initial consonant cluster of ON fr- to Tw (f->) h- or (r->) l- so that it forms Tw hū 婦 or lú 女. Through pleonastic formation (pairing of synonyms) of the two "daughter words", ON frú corresponds to Tw hū-lú 婦女.
(2) The Tw colloquial word for "pregnant" is pēnn-kiánn with pēnn being the nasalized form of pēng. The word pēng can be correlated with the first syllable praegn- of the Latin word praegnant (pregnant). Due to Tw phonology, the word-initial pr- is simplified to p-, and the word-stem final -gn is usally corresponded by Tw -ng, for example: ON hegn (I punish) : Tw hêng 刑 (punish). The Latin diphthong -ae- has been known to undergo a sound change to -e- in Romance languages. For example, L. praestare (to lend) becomes Italian prestare, French prêter, and Spanish, Portugese prestar.
Of all the words I have been able to find correspondents in European lexicons, about 50% are related to ON and other Germanics, and the other 50% to Latin-Greek. But that is another story.
Heruler
Exercise of PSC-1
In the back section of newspapers or magazines there are usually crossword puzzles, sudoku, chess games, etc to tease our brain. In the same spirit, let’s have some fun with PSC-1. Now that we have established the pattern of sound correspondence between Old Norse -angr and Taiwanese -an, we can apply this pattern to do an exercise.
I am going to give you an Old Norse word and ask you to find a corresponding word in Taiwanese/Hokkien, which should have a similar meaning and should follow the pattern of PSC-1.
ON krangr (weak, frail) : Taiwanese/Hokkien ___?___
Hint: How do you say in Taiwanese/Hokkien, “I am a little weak today”?
Note that the ON word krangr has a word-initial consonant cluster kr-. Because Taiwanese phonology does not allow a consonant cluster, we need to split it up first. This gives us two alternative leads for words with either k- or r- as the word-initial. In case you are pursuing the lead with the r- word-initial, remember to use a good substitute for r- because Taiwanese-Hokkien does not have the [r] sound.
I think this will be a rewarding experience for us all in that, not only we will have a little fun in solving a puzzle, but also we can draw satisfaction from seeing that PSC-1 is able to stand another test. So, let the game begin and please post your answer here.
Heruler
In the back section of newspapers or magazines there are usually crossword puzzles, sudoku, chess games, etc to tease our brain. In the same spirit, let’s have some fun with PSC-1. Now that we have established the pattern of sound correspondence between Old Norse -angr and Taiwanese -an, we can apply this pattern to do an exercise.
I am going to give you an Old Norse word and ask you to find a corresponding word in Taiwanese/Hokkien, which should have a similar meaning and should follow the pattern of PSC-1.
ON krangr (weak, frail) : Taiwanese/Hokkien ___?___
Hint: How do you say in Taiwanese/Hokkien, “I am a little weak today”?
Note that the ON word krangr has a word-initial consonant cluster kr-. Because Taiwanese phonology does not allow a consonant cluster, we need to split it up first. This gives us two alternative leads for words with either k- or r- as the word-initial. In case you are pursuing the lead with the r- word-initial, remember to use a good substitute for r- because Taiwanese-Hokkien does not have the [r] sound.
I think this will be a rewarding experience for us all in that, not only we will have a little fun in solving a puzzle, but also we can draw satisfaction from seeing that PSC-1 is able to stand another test. So, let the game begin and please post your answer here.
Heruler
Hi Heruler
This game is interesting. If traced to the r- initial, may be the word is 軟 luan2. Yet in my dialect we don't use luan2 in a context like "I am a little weak today".
The more common word would be 倦 sian7. 倦 is juan4 in Mandarin, and based on "sound-shifting" it should sound something like kuan7/3 in Hokkien (literary reading), which may correspondent to "krangr" [by tracing via initial k-]. I am not aware what's the correct literary reading for 倦 though.
Btw, I heard that Taiwanese usually use thiam2 instead of sian7. The former in our dialect has more severe meaning, closer to 慘 cham2.
This game is interesting. If traced to the r- initial, may be the word is 軟 luan2. Yet in my dialect we don't use luan2 in a context like "I am a little weak today".
The more common word would be 倦 sian7. 倦 is juan4 in Mandarin, and based on "sound-shifting" it should sound something like kuan7/3 in Hokkien (literary reading), which may correspondent to "krangr" [by tracing via initial k-]. I am not aware what's the correct literary reading for 倦 though.
Btw, I heard that Taiwanese usually use thiam2 instead of sian7. The former in our dialect has more severe meaning, closer to 慘 cham2.
Hi niuc,
Let me submit my answer first and then discuss the various possibilities.
My answer is: lán (懶). “I am a little weak today” can be translated as “Kin-á jít góa ū sió khóa lán-lán.” “Góa ê pēnn ū khah hó à, put-kò iáu sió khóa lán-lán leh” means “My illness is getting better, but I am still feeling a little weak.”
Your answer loán 軟 is also right on the mark. Since it is common to use 2 synonyms together to form a compound word with the same meaning (pleonastic), loán-jiók 軟弱 means ‘weak’, and so is jiók 弱 . Then loán 軟 must also mean ‘weak’, even though we don’t use it for the expression such as “I am a little weak today”. Sorry, my hint seemed to make things more difficult for you.
In the Bible, Acts 3: 2, “a man crippled from birth” is translated by Rev. Barclay into Taiwanese as “有一人, 對出世就軟腳 Ū chít lâng, tùi chhut-sì chiū nńg kha.” Here the vernacular term nńg kha 軟腳 is a very apt translation for ‘cripple’ since muscle disuse usually results in muscle weakness.
Because Taiwanese/Hokkien very often turns a word-final -n to -m, the word lán 懶 has a sister-word lám. To my knowledge, lám does not seem to have a Chinese cognate, so I would tentatively classify it as a Group-1 word. This word applies to humans, animals as well as inanimate objects. Examples: lám-lâng (a weak person), lám-bé (a weak horse). “Chit-tè tek-á chò ê í-á sī khah lám” means “This chair made of bamboo is rather weak.” There is an idiom, “Lám-lám bé mā ū chít pō· that” which means “A weak horse can still give a good kick”, a saying often used to encourage people.
So, from ON krangr 'weak, frail', we are able to find 3 corresponding words, two of which conform to PSC-1, and the third one has undergone further sound change.
Now, there is a little complication. Because the Taiwanese/Hokkien vocabulary has a lot of words that show sound correspondences to Latin, I would like to point out one for lán / loán. Latin languor ‘disease, faintness, weariness’ gave rise to the Romance word *languire from which the English word languish is derived. Languish means ‘to grow weak or faint’. So, from the first syllable of *languire we can derive Taiwanese/Hokkien lán. Could it be that lán 懶 is related to Latin languor while loán 軟 is related to ON krangr? I don’t know.
Footnote: koàn (倦), siān, thiám: the three words you mentioned denote “tired, getting tired”, which are somewhat different from “weak, frail”. I agree with you that it’s tempting to see koàn 倦 as a corresponding word for ON krangr.
Heruler
Let me submit my answer first and then discuss the various possibilities.
My answer is: lán (懶). “I am a little weak today” can be translated as “Kin-á jít góa ū sió khóa lán-lán.” “Góa ê pēnn ū khah hó à, put-kò iáu sió khóa lán-lán leh” means “My illness is getting better, but I am still feeling a little weak.”
Your answer loán 軟 is also right on the mark. Since it is common to use 2 synonyms together to form a compound word with the same meaning (pleonastic), loán-jiók 軟弱 means ‘weak’, and so is jiók 弱 . Then loán 軟 must also mean ‘weak’, even though we don’t use it for the expression such as “I am a little weak today”. Sorry, my hint seemed to make things more difficult for you.
In the Bible, Acts 3: 2, “a man crippled from birth” is translated by Rev. Barclay into Taiwanese as “有一人, 對出世就軟腳 Ū chít lâng, tùi chhut-sì chiū nńg kha.” Here the vernacular term nńg kha 軟腳 is a very apt translation for ‘cripple’ since muscle disuse usually results in muscle weakness.
Because Taiwanese/Hokkien very often turns a word-final -n to -m, the word lán 懶 has a sister-word lám. To my knowledge, lám does not seem to have a Chinese cognate, so I would tentatively classify it as a Group-1 word. This word applies to humans, animals as well as inanimate objects. Examples: lám-lâng (a weak person), lám-bé (a weak horse). “Chit-tè tek-á chò ê í-á sī khah lám” means “This chair made of bamboo is rather weak.” There is an idiom, “Lám-lám bé mā ū chít pō· that” which means “A weak horse can still give a good kick”, a saying often used to encourage people.
So, from ON krangr 'weak, frail', we are able to find 3 corresponding words, two of which conform to PSC-1, and the third one has undergone further sound change.
Now, there is a little complication. Because the Taiwanese/Hokkien vocabulary has a lot of words that show sound correspondences to Latin, I would like to point out one for lán / loán. Latin languor ‘disease, faintness, weariness’ gave rise to the Romance word *languire from which the English word languish is derived. Languish means ‘to grow weak or faint’. So, from the first syllable of *languire we can derive Taiwanese/Hokkien lán. Could it be that lán 懶 is related to Latin languor while loán 軟 is related to ON krangr? I don’t know.
Footnote: koàn (倦), siān, thiám: the three words you mentioned denote “tired, getting tired”, which are somewhat different from “weak, frail”. I agree with you that it’s tempting to see koàn 倦 as a corresponding word for ON krangr.
Heruler
Hi Heruler
Thank you for the detailed analysis and good samples. I am not a linguist and don't have much knowledge about European languages, so I can't really contribute much.
In my dialect, we never use lan2 (lan2-lan2) such as in your example "Kin-á jít góa ū sió khóa lán-lán". Instead, we use lam2 to replace lan2-lan2 in the examples. Btw 懶 is also lam2, as in lam2-nua*7 懶惰, right? But 懶 lam2 here means lazy instead of weak. Although 弱 may not have lam2 as a reading, nonetheless the meaning (weak) is just right.
軟 luan2/nng2 primarily means "soft" in our usage, and by extension "weak" [such as 軟腳 nng2-kha and nng2-cia*2].
Thank you for the detailed analysis and good samples. I am not a linguist and don't have much knowledge about European languages, so I can't really contribute much.
In my dialect, we never use lan2 (lan2-lan2) such as in your example "Kin-á jít góa ū sió khóa lán-lán". Instead, we use lam2 to replace lan2-lan2 in the examples. Btw 懶 is also lam2, as in lam2-nua*7 懶惰, right? But 懶 lam2 here means lazy instead of weak. Although 弱 may not have lam2 as a reading, nonetheless the meaning (weak) is just right.
軟 luan2/nng2 primarily means "soft" in our usage, and by extension "weak" [such as 軟腳 nng2-kha and nng2-cia*2].
Hi Heruler,
Long time no see - sorry for not answering earlier, the last couple of weeks have been extremely busy. For the same reason, I will only give a very short reply today, my apologies.
Thanks for explaining your working model in another thread. As far as data collection goes, I am very happy to hear a fellow scientist speak. I'd caution, though, that increasing the sample size alone does not guarantee validity of the results. Let me give an example from biology: If you try to identify proteins based on their mass spectra and feed the signals you get into a database, you'll have to specify how much deviation between the measured masses and the database masses you're willing to accept - any instrument you use will have a bit of measurement error, so if you set the allowable deviation to zero, you will never get a match. If you make the allowable deviation too large, though, you'll get tens of thousands of potentially identified proteins. In other words: If you make your acceptance criteria too weak, you'll get a lot of false positives.
How does this relate to the case at hand? Let me give you two examples (sorry, so little time):
亂 loan's basic meaning clearly is not 'wrong' but 'chaos' like in the Kurosawa movie (Ran) or when your mom says that your room is cin loan (or luan4 qi1 ba 1 zao1) ...
姦 means licentiousness and a couple of other things all related to libido ... the second syllable in the Germanic compound (gangr) clearly comes from the a verb for 'to go' like in German 'ich gehe, ich ging, war ge-gang-en ... now, I'm fairly certain the character in question never occurs with reference to 'walking' or the like, so that's clearly a false positive
I'd advise against stretching the meanings of words too much because this will produce too many false matches.
Likewise, I'd also counsel against stretching the acceptance criteria for pronunciation too much: gardr losing the final cluster 'rdr' - even if we ignore the nominative ending -r, we have just deleted 50% of the phonetic information of the word - that is exactly what I meant by my protein example - if I make the acceptance criteria wide enough, anything will fit through. Moreover, I believe the meaning of gard-r is more that of an enclosure/ farm - isn't the common word for house something like hus? Moreover, 家 is not the common word for house but for family, at least in contemporary usage ... certainly one could imagine the change in meaning AND pronunciation if there was other proof ... but as 'exhibit A' this seems weak
On a side note: In the early days of Western etymology (which WAS dreadful at the time), somebody remarked wittily that 'the etymologists do not care much about the vowels and do not pay attention to the consonants' - let's not make that mistake again ...
Best,
Aurelio
Long time no see - sorry for not answering earlier, the last couple of weeks have been extremely busy. For the same reason, I will only give a very short reply today, my apologies.
Thanks for explaining your working model in another thread. As far as data collection goes, I am very happy to hear a fellow scientist speak. I'd caution, though, that increasing the sample size alone does not guarantee validity of the results. Let me give an example from biology: If you try to identify proteins based on their mass spectra and feed the signals you get into a database, you'll have to specify how much deviation between the measured masses and the database masses you're willing to accept - any instrument you use will have a bit of measurement error, so if you set the allowable deviation to zero, you will never get a match. If you make the allowable deviation too large, though, you'll get tens of thousands of potentially identified proteins. In other words: If you make your acceptance criteria too weak, you'll get a lot of false positives.
How does this relate to the case at hand? Let me give you two examples (sorry, so little time):
亂 loan's basic meaning clearly is not 'wrong' but 'chaos' like in the Kurosawa movie (Ran) or when your mom says that your room is cin loan (or luan4 qi1 ba 1 zao1) ...
姦 means licentiousness and a couple of other things all related to libido ... the second syllable in the Germanic compound (gangr) clearly comes from the a verb for 'to go' like in German 'ich gehe, ich ging, war ge-gang-en ... now, I'm fairly certain the character in question never occurs with reference to 'walking' or the like, so that's clearly a false positive
I'd advise against stretching the meanings of words too much because this will produce too many false matches.
Likewise, I'd also counsel against stretching the acceptance criteria for pronunciation too much: gardr losing the final cluster 'rdr' - even if we ignore the nominative ending -r, we have just deleted 50% of the phonetic information of the word - that is exactly what I meant by my protein example - if I make the acceptance criteria wide enough, anything will fit through. Moreover, I believe the meaning of gard-r is more that of an enclosure/ farm - isn't the common word for house something like hus? Moreover, 家 is not the common word for house but for family, at least in contemporary usage ... certainly one could imagine the change in meaning AND pronunciation if there was other proof ... but as 'exhibit A' this seems weak
On a side note: In the early days of Western etymology (which WAS dreadful at the time), somebody remarked wittily that 'the etymologists do not care much about the vowels and do not pay attention to the consonants' - let's not make that mistake again ...
Best,
Aurelio
Hi Aurelio,
Really long time no see! Glad to hear from you!
I agree with your caution regarding the acceptable level of statistical deviation. But generally, the central theorem tells us that the more samples you get, the closer you are to the true value. This is what we normally do in biology. In my research I deal with stationary autocorrelation and fluctuation analysis.
Regarding the word 亂 loān, it has more than one meaning (chaos). It also means 'not according to the normal, right way' (不按照常理做事) in Chinese as well as in Taiwanese/Hokkien. For examples, 亂講 loān-kóng (= not saying the right thing), 亂來 loān-lâi (= not doing the right thing). Thus, 亂判 loān-phoànn (= not judging the right way) fits this connotation.
I agree the Chinese character 姦 has the meaning of licentiousness, etc. In Taiwanese kàn is of the third tone whereas 姦 has the first tone. I followed Campbell's dictionary which uses 姦 in parenthesis for kàn, i.e., borrowing literary character for colloquial usage. I should have put a footnote there when I cited it.
It's true that the second syllable in sam-gangr (gangr) means walking. As in German, Ausgang & Eingang. In Old Norse, sam-gangr has several meanings: (1) going together; (2) intercourse; (3) marriage; (4) conflict, fight (Reference: Zoëga's A Concise Dictionary of Old Icelandic, page 347). I cited the second definition for intercourse. Interestingly, in Taiwanese, when a man is going together with a woman in courtship, we say, sann-kiânn, 'going together'. Two matches ('going together' & 'intercourse') for one word, like two birds with one stone! However, the kiânn follows another pattern of sound correspondence.
Regarding 家 ka, or with sound change, ke, the Old Norse corresponding word garðr has several closely related meanings: (1) fence, wall; (2) enclosed space, yard; (3) court-yard; (4) house, dwelling; and (5) stronghold, castle. Obviously, meaning No. 4 fits the Taiwanese/Hokkien word. The usage of 家 ka for house is well established. Recall a poem by Chhui Hō 崔顥 from the Tang Dynasty: "君家何處住? 妾住在橫塘." Kun ka hô chhù chū? Chhiap chū chāi Hêng-tông (Where is thy house? I live in Heng-Tang.)
You mentioned house. Ha-ha!!
The Old Norse word hús has a Taiwanese/Hokkien corresponding word, hú 府. This word is a very polite term for 'house'. If we translate the sentence, "請問貴府在何處?" into English with the right tone, it probably goes like this, "Where is thy mansion, may I enquire?" In Taiwan, the Presidential House is called, Chóng-thóng Hú 總統府, highly appropriate for dignity.
Now, there is a pattern of sound correspondence (PSC) between ON and Taiwanese/Hokkien of -u- > -u-. Examples:
hús 'house' : hú 府 'house' (polite term) : German Haus
húð 'skin, hide' : hu 膚 'skin' : German Haut
frú 'lady' : hū 婦; lú 女 'lady' : German Frau
brugg 'brewing' : phū (G1W) 'brewing' : German brauen
tún 'enclosure, fence, town' : tùn 屯 'town, village' (German Zaun 'fence' only)
In the last example, only Norse and English developed the meaning from 'fence' to 'town' (Kluge's Etymologische Wörterbuch). Examples of English placenames with -ton (derived from OE tūn = ON tún) are numerous: Boston, Washington, Houston, Lexington, Barrington...
In Taiwan, there are many placenames with tùn 屯 such as 沙屯, 白屯, 西屯, 南屯, 屯仔腳... Thus, the Taiwanese usage of tùn fits the ON or OE usage. (By the way, in Japanese, 屯 is pronounced similar to English, ton. And a Japanese etymological dictionary 漢字源 even states that this Japanese word may have something to do with the English ton.)
When we compare Old Norse, Taiwanese/Hokkien and New High German words in this PSC, we see that the former two are very closely related, whereas German has undergone a sound change to -au-.
I enjoy chatting with you, especially with your German backgrounds.
Heruler
Really long time no see! Glad to hear from you!
I agree with your caution regarding the acceptable level of statistical deviation. But generally, the central theorem tells us that the more samples you get, the closer you are to the true value. This is what we normally do in biology. In my research I deal with stationary autocorrelation and fluctuation analysis.
Regarding the word 亂 loān, it has more than one meaning (chaos). It also means 'not according to the normal, right way' (不按照常理做事) in Chinese as well as in Taiwanese/Hokkien. For examples, 亂講 loān-kóng (= not saying the right thing), 亂來 loān-lâi (= not doing the right thing). Thus, 亂判 loān-phoànn (= not judging the right way) fits this connotation.
I agree the Chinese character 姦 has the meaning of licentiousness, etc. In Taiwanese kàn is of the third tone whereas 姦 has the first tone. I followed Campbell's dictionary which uses 姦 in parenthesis for kàn, i.e., borrowing literary character for colloquial usage. I should have put a footnote there when I cited it.
It's true that the second syllable in sam-gangr (gangr) means walking. As in German, Ausgang & Eingang. In Old Norse, sam-gangr has several meanings: (1) going together; (2) intercourse; (3) marriage; (4) conflict, fight (Reference: Zoëga's A Concise Dictionary of Old Icelandic, page 347). I cited the second definition for intercourse. Interestingly, in Taiwanese, when a man is going together with a woman in courtship, we say, sann-kiânn, 'going together'. Two matches ('going together' & 'intercourse') for one word, like two birds with one stone! However, the kiânn follows another pattern of sound correspondence.
Regarding 家 ka, or with sound change, ke, the Old Norse corresponding word garðr has several closely related meanings: (1) fence, wall; (2) enclosed space, yard; (3) court-yard; (4) house, dwelling; and (5) stronghold, castle. Obviously, meaning No. 4 fits the Taiwanese/Hokkien word. The usage of 家 ka for house is well established. Recall a poem by Chhui Hō 崔顥 from the Tang Dynasty: "君家何處住? 妾住在橫塘." Kun ka hô chhù chū? Chhiap chū chāi Hêng-tông (Where is thy house? I live in Heng-Tang.)
You mentioned house. Ha-ha!!
The Old Norse word hús has a Taiwanese/Hokkien corresponding word, hú 府. This word is a very polite term for 'house'. If we translate the sentence, "請問貴府在何處?" into English with the right tone, it probably goes like this, "Where is thy mansion, may I enquire?" In Taiwan, the Presidential House is called, Chóng-thóng Hú 總統府, highly appropriate for dignity.Now, there is a pattern of sound correspondence (PSC) between ON and Taiwanese/Hokkien of -u- > -u-. Examples:
hús 'house' : hú 府 'house' (polite term) : German Haus
húð 'skin, hide' : hu 膚 'skin' : German Haut
frú 'lady' : hū 婦; lú 女 'lady' : German Frau
brugg 'brewing' : phū (G1W) 'brewing' : German brauen
tún 'enclosure, fence, town' : tùn 屯 'town, village' (German Zaun 'fence' only)
In the last example, only Norse and English developed the meaning from 'fence' to 'town' (Kluge's Etymologische Wörterbuch). Examples of English placenames with -ton (derived from OE tūn = ON tún) are numerous: Boston, Washington, Houston, Lexington, Barrington...
In Taiwan, there are many placenames with tùn 屯 such as 沙屯, 白屯, 西屯, 南屯, 屯仔腳... Thus, the Taiwanese usage of tùn fits the ON or OE usage. (By the way, in Japanese, 屯 is pronounced similar to English, ton. And a Japanese etymological dictionary 漢字源 even states that this Japanese word may have something to do with the English ton.)
When we compare Old Norse, Taiwanese/Hokkien and New High German words in this PSC, we see that the former two are very closely related, whereas German has undergone a sound change to -au-.
I enjoy chatting with you, especially with your German backgrounds.
Heruler
Sorry to be slightly off topic here. I learnt the characters 懶惰 for "lazy" in Mandarin just a few weeks agoniuc wrote:...Btw 懶 is also lam2, as in lam2-nua*7 懶惰, right? But 懶 lam2 here means lazy instead of weak... [such as 軟腳 nng2-kha and nng2-cia*2].
. I had never heard this Hokkien equivalent "lam2-nua*7" before. We always say "pun7/3-tua*7/3" Sim.
P.S. My usual apologies for the 7/3 business... Tone 7 and tone 3 sound the same to me, so the only way I can tell them apart is "the one which changes sandhi-tone is the 3, and the one which doesn't is the 7"! I don't know "pun" and "tua*" in any other combination than "pun-tua", so I can't tell!
