| View previous topic :: View next topic |
Donmillion
In: Acton, Middlesex
|
|
 |
|
D.P. Crisp wrote:
| Donmillion, do you know of any resources that actually demonstrate the written evidence of the Great Vowel Shift? I haven't found a single one that doesn't just assert it using the IPA. |
Well, Dan, problem is, there's a gotcha: 'The spelling of most English vowels is frozen at a form which precedes the Great Vowel Shift' (Kenschaft, http://tinyurl.com/35nkbul ). Therefore, you can't "see" the GVS occurring via a change in spelling.
However, that itself is the principal evidence for change: 'This explains why English vowel symbols no longer correspond to the sounds they once represented in English and still represent in other modern languages'.
Absorbed as we are in the cradle-to-grave conventions of our own culture, it's difficult for English speakers to appreciate how bizarrely our vowel orthography departs from the norm for users of the Roman alphabet. But that's where the evidence for change is found: in, for example, English spelling and pronunciation of words from Old French (or Anglo-Norman French).
Here are several examples, which I'll try to illustrate by using modern English vowel sounds to approximate the Old French pronunciations (which will be a strain at some points, and I may have slipped at some). For the most part, Old French spelling was essentially the same as the modern English spelling used in the list. Where there is a substantial difference, I've shown the OF spelling:
aid: Old French /edd/
beast: Old French beste /best/
cry: Old French crier /cree-air/
dime: Old French /deem/ /
estate: Old French estat /estaht/
flower: Anglo-Norman flur /floo-r/
grace: Old French /grahss/
ivory: Old French yvoire /eevwahr/
libel: Old French /leebell/
mail: Old French male /mahl/
nice: Old French /neess/
poach: Old French pochier /poshyair/
rage: Old French /rahzh/
save: Anglo-Norman sa(u)ver /sahwvair/
taste: Old French tast /tahst/
viol: Old French vielle /vee-ell/
wage: Anglo-Norman /wahzh/
-- and any word ending in -ation (French /-assyon/ versus English /-eishun/).
Of course, to believe what this list purports to tell you, you have to believe that Old French and Anglo-Norman French were pronounced pretty much as spelt, and using 'continental' vowel values. Since the essential values of the monophthong vowels a-, e-, i-, o-, and u- are the same in almost all Continental languages, this is taken to be true; but of course could be disputed.
It is not necessary to believe that any of the words was derived from (Old) French, only that modern English spelling reflects the pronunciations used in Old French, and therefore represents the pronunciations used in Old French.
One final point: it is possible (under the 'what is now is what was' principle) to dispute the notion that our modern spellings reflect an older, pre-GVS, pronunciation system for vowels, but I think it's harder in the case of consonants. Why are men who clank around in armour 'knights' and not 'nites'? And why is -ough pronounced six different ways, none of them involving a 'g' sound?
Of course it can be argued that those spellings don't indicate sound change; perhaps the spelling 'knight' was adopted from German Knecht, with a couple of minor changes, despite the fact that the spelling was nothing like the English pronunciation. But on the whole, I think that's unlikely, and I think the idea that English chose a totally different way of representing vowel sounds from anyone else in Europe is equally unlikely.
|
|
  |
|
 |
|
|
|
|
|
|
| Donmillion wrote: | frank h wrote:
English -by in place-names reflects Old Norse byr, basically meaning 'a dwelling' ..... the Normans were Danish (should look it up, really), given the very large number of toft and especially thorp names, including Yvetot, Lilletot ('Little Toft'?), Hautot and Langtot ('high' and 'long'), and tourp, torbe, and tourbe names, not to mention "Le Torpe". |
Presumably the Normandy 'by' places are related to the German word bau = building or bauer = farmer. And 'thorpe' to dorf = village. The latter being common across Germania and occur more or less in the same regions as the 'by' places in Britain.
It would be interesting to know if any 'by' and 'thorpe' places exist in the Ukraine around Kiev or down the east coast of Ireland, since these would perhaps be better pointers to 9th century Viking activity for the place name.
|
|
| |
|
|
|
Donmillion
In: Acton, Middlesex
|
|
|
|
D.J. Crisp wrote (quoting me):
| Our cousins John and Mary are cognates because they also have the same parents (as one another, not as us). And the four of us are cognates because we all have the same pair of grandparents |
If you can turn down the analyticity, megaDon, you should find all this a piece of piss. In your example, you and your cousins are clearly not cognates; and at the same time, in another sense, you clearly are. Keep hold of such Fuzzy Logic concepts. |
Superposition of states, eh? 'Cognate' and 'not cognate' at the same time!
But I can't agree. The essential definition of cognate is, 'descended from a common ancestor', which all of us (me, my sister, my cousins) 'clearly' are--otherwise we wouldn't be cousins. The secondary definition (one I wasn't really aware of until a few days ago) is, 'Akin in origin, nature, or quality'. Clearly we're akin in origin (we're 'kin'), and we're also akin in nature (we're human beings) and quality (we're all intelligent adults of 'lower middle class' status).
Are you appealing to the literal sense of 'cognate'? It's true that my cousins and I were not 'born together', but the literal sense of the elements of a word don't necessarily indicate the actual meaning of the word, and the various on-line Latin dictionaries I've been able to consult all refer to blood kinship without necessary co-parentage.
So would you please explain, Dan (I'm obviously not fuzzy enough yet) in what sense I and my cousins are not cognates?
|
|
| |
|
|
|
Ishmael
In: Toronto
|
|
|
|
| Donmillion wrote: | | The essential definition of cognate is, 'descended from a common ancestor'... |
We don't use the term that way.
We do not recognize any methodology by which ancestors can be differentiated from their offspring.
|
|
| |
|
|
|
Donmillion
In: Acton, Middlesex
|
|
|
|
D.J. Crisp wrote:
| We all agree (?) that languages evolved from ancestral forms and equivalent words in different languages range from being identical... through being pronounced/spelled their respective ways... to settling in different parts of a spectrum of meaning. |
Well, I do at least.
| We typically distinguish loanwords as being fairly consciously used as foreign words, even though "born together" clearly applies. |
Let me deal with 'loanwords being fairly consciously used as foreign words' first. Look at the words in my response re the GVS, all of which are reputedly loan-words into English from Old French (aid, beast, mail, poach, taste ...). I argue that few English speakers would think of any of those words as 'foreign words', except those who've been taught so in French lessons. But they were all adopted into English (as far as we can tell), following the Norman Conquest.
However: are they cognates, and if so, of what--one another, English words, or both?
Recall that, whatever the meanings of its respective Latin roots (giving 'born together'), 'cognate' is defined as meaning 'descended from a common ancestor' (i.e., not restricted to a common parent, which is why my cousins and I are cognates). This is also its meaning in Latin. Now, your assertion is stated as a universal. For it to be universally true requires that all words adopted into any one language from any other languages share a common origin either with one another, or with words in the adopting language, or both.
Now, this might be true (that loanwords are cognates of one another or of words in the adopting language). But while I am excited by the possibilities of Nostratic studies, I'm far from convinced as yet by their results.
There are lots of Maori loanwords in common use in New Zealand English, such as kea (a type of parrot) or koha (a donation). These two words may be cognates of one another (in the dictionary sense of deriving from a common ancestral word in Maori or some other language), but I cannot see the 'family relationship' except in the most general terms. On the same basis, you and I are cognates because (I'm assuming here!) we have the same numbers of bodies, arms, legs, and heads as one another. By the same token, we're also cognates of cats, dogs, and horses, and (if you extend 'arms' to include 'wings') budgerigars and eagles (perhaps of half an octopus, too ...).
Not only do the Maori words not appear to be cognates of one another, they also are not apparent cognates of any words in their host language, English. Past 'histories' of Maori and English words could be invented so as to take them back to a common 'ancestor'; for example, the hypothetical word *glako developed into *kleako in some hypothetical ancestor-language of Maori, which developed into *keako and, eventually, kea. Meanwhile, in parallel with this process of development, *glako developed into *glaro in some ancestor of English, which developed into *splaro and eventually wound up as sparrow. So kea and sparrow would be cognate words.
But there's no evidence at all for any such common ancestry. Until there is, kea must be regarded as a Maori loanword in English, not cognate with any English word, and likewise not cognate with koha. 'Born together' certainly does not apply. To the fullest extent of our current knowledge, kea is part of the English word family by adoption, not by birth.
We refer to cognate sounds because they have Family Resemblances, too: they're generated in similar ways, are easily mispronounced or mistaken for one another and so on.
And we refer to letters as cognates when they represent cognate sounds, which is sometimes reflected in their shapes. The relationship between letters and sounds lies at the heart of the problem with historical linguistics. |
Thanks for this clarification. I'm not comfortable with it, because it means this community, or rather individual elements within this community, use 'cognate' in at least two quite different senses, which will inevitably cause confusion. In fact, I suspect that you (individuals within the AE community) are using 'cognate' in only one sense, but one that is infinitely flexible to take in any meaning you want it to: 'any apparent relationship whatsoever'.
Whereupon mutual understanding and reasoned discussion become impossible. This is just about confirmed by Ishmael's most recent comment: Cognate means 'descended from a common ancestor'?
| We don't use it that way. |
How do you use it? And how many of us are included in 'we'?
Hair-tearing time. Let me declare right now that 'cognate' actually means 'having no relationship at all, because it is impossible to identify relationships when there is no 'methodology by which ancestors can be distinguished from their offspring'.
Sorry, guys. You have convinced me that 'cognate' is the wrong word for what you want to express, unless your specific purpose is to sow as much confusion as possible so as to make answers impossible. I think the term you want instead of 'cognate' is, 'possibly vaguely related in some way because of a linguistic, phonetic, semantic, or orthographic resemblance of some sort, for which I can invent an unsubstantiated explanation'.
I'm afraid Ishmael has just excluded himself from reasoned discourse. Sorry, Ishmael.
|
|
| |
|
|
|
Mick Harper
Site Admin
In: London
|
|
|
|
As a general guide, Dan is a good guide, Ishmael less so. I cannot believe anybody would waste ink on the meaning of 'cognate' in linguistics. It is used to identify any word that has the same approx sound and meaning in two related languages ie a common source accounts for the resemblance. When the same thing happens in unrelated languages, then the presumption is that one is loaned from the other.
It is true that a word that would ordinarily be considered as a cognate turns out, on correct analysis, to be loaned from a related language. There are various other possible complications too eg the word was cognated and then lost in the language that subsequently loaned it back! etc etc. And, as we all know, the biggest complication of all is the sheer unlikelihood (and mostly downright impossibility) of correct analysis.
|
|
| |
|
|
|
Ishmael
In: Toronto
|
|
|
|
| Mick Harper wrote: | | As a general guide, Dan is a good guide, Ishmael less so. |
I concur.
|
|
| |
|
|
|
Donmillion
In: Acton, Middlesex
|
|
|
|
Mick wrote:
| All this arises from my initial perusal of a Loeb's Classical Edition of something or other when the English text is faced by the Latin text, and the Latin takes up approx half the space. For page after page after page. Since English takes up much the same space as all other (modern, natural European) languages, this points to Latin as being expressly compressed. |
You called me 'Donna'. That's Prima!
I'd like to know what text it was, and what translation. In English, two different writers can say the same thing in quite different numbers of words, and a Victorian-era translator is likely to have used a quite different vocabulary from what a modern translator would use, and possibly also a lot more punctuation (which can take up a surprising amount of space). By the same token, some Latin authors deliberately wrote in a very compressed style, which their contemporaries hailed as 'elegant' but which readers often found difficult to follow.
It's quite possible, therefore, that you were looking at a very concise Latin author afflicted with a very prolix English translator. But since I can't see the volume you cite, that's speculation. Let's move to the concrete.
A partial explanation for the 'compression' is the reduction in spaces in Latin, because English (and Italian) uses separate words for grammatical functions that Latin indicates with suffixes. For example, if we used Latin-style affixes instead of independent particles, English 'from the mouth of the horse' might be represented as 'fromthemouth ofthehorse'--an instant reduction from 27 characters to 23. If you're right about a 50% compression ratio, there's 15% for a start.
Remember, too, that when they invented Latin, the Italians dropped the definite article, so I should really have written, 'frommouth ofhorse' only 17 characters, so a reduction by 37%.
The first four sentences of Caesar's Gallic Wars total 490 characters in Latin, including spaces (and punctuation). A smooth translation into standard modern English has 565 characters: 15% more than the Latin. Removing definite and indefinite articles (as the Italians did), but still counting all spaces, reduces the English text to 511 characters, only 4% longer than the Latin.
If I tack prepositions onto the front ends of nouns, and pronouns and aspect markers onto the front of verbs, instead of representing them as separate words, then removing the spaces further reduces the English text to a mere 496 characters, which is only 1% longer than the Latin. And part of that difference comes from the fact that, in a few cases, Latin has quite short words where English has a longer one--Latin 'se' for English 'themselves', for example. (If there were a two-letter English word for 'themselves', the 'Latinised' English text would be slightly shorter than the Latin.)
Hmm. Simply by removing some spaces and all occurrences of 'the' and 'an', we've reduced the English text by 12%; and once you get used to prepositions, etc, being prefixed to words (rather than suffixed as in Latin),with no intervening space, it's still very readable:
| All Gaul is divided in three parts: one ofwhich Belgae inhabit; other Aquitani; third, those who intheirown language arecalled Celts, inours, Gauls. All these differ between themselves inlanguage, institutions, laws. River Garumna divides Gauls from Aquitani, Marne and Seine from Belgae. Belgae are bravest ofall these: because theyare mostdistant from cultivation and refinement ofProvence; and merchants leastoften resort to them, and import thosethings which appertain to effeminate minds. |
Here's Caesar's Latin, for those who might want to check what I've done:
| Omnia Gallia est divisa in tres partes: unum quarum Belgae incolunt; aliam Aquitani; tertiam, qui lingua ipsorum appellantur Celtae, nostra, galli. Omnes hi different inter se lingua, institutis, legibus. Flumen Garumna dividit Gallos ab Aquitanis, Matrona et Sequana a Belgis. Belgae sunt fortissimo omnium horum: propterea quodabsunt longissime a sultu atque humanitate Provinciae; que mercatores minime seape commeant as eos, atque important ea, quae pertinent ad animos effeminandos. |
Sorry, Mick, I don't see that adding thousands of grammatical rules to achieve the same effect as I've achieved by removing articles and a few spaces is worth the effort.
With regard to the effort in reducing and transliterating 'the three hundred and seventy eight sounds of natural Campanian/Italian of c 600 BC -- (how do you know that? Pre-Classical sound recordings? Time machine?)--I'm afraid you're confusing phonetic transcription with phonemic transcription. Well. Not quite--I know you're not thinking of the phonetic transcription of 378 vocal sounds. Be that as it may, phoneticists regularly use three forms of 'phonetic' transcription:
narrow phonetic transcription, which would try to capture the 378 vocal noises you mention;
broad phonetic transcription, which would try to capture the much smaller number (typically around fifty) that may make a difference in a native speaker's understanding of words; and
phonemic transcription, which aims to capture the smaller number still, perhaps around forty, that native speakers are conscious of.
Phonetic transcriptions are normally placed in [square brackets], phonemic transcriptions between /slashes/.
I've no idea how many phonemes the 600 BC Italians actually had, but I doubt that they were aware of the minutely subtle phonetic difference between c followed by a and c followed by e. Can you hear any difference in the initial consonants of 'cat' and 'kettle'? I strongly doubt it, unless you've been trained in what to listen for. But the following vowel marginally changes the exact position of the tongue against the palate, which produces a subtle difference that is quite obvious if you see the wave-forms oscilloscoped, and which eventually led to Latin ci- words being pronounced /si-/ in French and /chi-/ in Italian (or so the story goes).
Those are the sorts of subtle differences that account for your 'three hundred and seventy eight sounds', Mick, and the great majority of them are totally inaudible to untrained ears and--more to the point--have absolutely no significance when it comes to the meaning of the words.
An example from Italian: Neapolitan speakers place a tiny 'kh' sound (like the -ch of Scottish 'loch') at the end of the word si ('yes'). They're normally unaware of it, and it makes no difference to the meaning of the word. A phonemic transcription would give you /si/, and a broad phonetic transcription ('capturing what they said') would be the same but in different brackets, [si]. But a narrow transcription captures "how they said it", including the tiny 'kh' sound represented via a superscript [x] character that I can't properly represent here: [six].
| But then nobody ever had to speak this new language...did they? |
Ever heard of Plautus? Terence? Seneca?
|
|
| |
|
|
|
Donmillion
In: Acton, Middlesex
|
|
|
|
As an experiment, I went through my "Latinised translation" of Caesar, and to the best of my ability substituted "Old English"-derived words for Latin-derived ones. Here's the result:
| All Gaul is split in three bites: one ofwhich Belgae livein; other Aquitani; third, those who intheirown tongue arenamed Celts, inours, Gauls. All these warp between themselves intongue, worldthews, setnesses. Stream Garumna sunders Gauls from Aquitani, Marne and Seine from Belgae. Belgae are bravest ofall these: because theyare mostfar from bigging and refinement ofProvence; and sellers leastoften go to them, and bringin thosethings which belong to womanly minds. |
That's 466 characters, including spaces: a 5% reduction from the Latin!
_________________
-- Don
"Eveything is deeply intertwingled" (thankyou, Danny Faught)
|
|
| |
|
|
|
Mick Harper
Site Admin
In: London
|
|
|
|
But then nobody ever had to speak this new language...did they?
Ever heard of Plautus? Terence? Seneca? |
But then nobody ever had to speak this NEW language...did they?
|
|
| |
|
|
|
Donmillion
In: Acton, Middlesex
|
|
|
|
'When I use a word', Humpty Dumpty said, in rather a scornful tone, 'it means just what I choose it to mean--neither more nor less'.
'The question is', said Alice, 'whether you can make words mean so many different things.'
'The question is,' said Humpty Dumpty, 'which is to be master--that's all.'
Alice was too much puzzled to say anything; so after a minute Humpty Dumpty began again. 'They've a temper, some of them--particularly verbs: they're the proudest -- adjectives you can do anything with, but not verbs -- however, I can manage the whole lot of them! Impenetrability! That's what I say!'
'Would you tell me please,' said Alice, 'what that means?'
'Now you talk like a reasonable child,' said Humpty Dumpty, looking very much pleased. 'I meant by 'impenetrability' that we've had enough of that subject, and it would be just as well if you'd mention what you mean to do next, as I suppose you don't mean to stop here all the rest of your life.'
'That's a great deal to make one word mean,' Alice said in a thoughtful tone. |
Mick wrote,
| I cannot believe anybody would waste ink on the meaning of 'cognate' in linguistics. It is used to identify any word that has the same approx sound and meaning in two related languages ie a common source accounts for the resemblance. When the same thing happens in unrelated languages, then the presumption is that one is loaned from the other. |
To quote the humble Ishmael, 'I concur'. BUT ...
The reason, Mick, that so much 'ink' has been spilt on it in the past few days is quite simple: Your description of 'cognate', above, is quite different from your description in THOBR that I quoted at the start of all this (--'Cognate' words are either the same as those in another language or corrupted or evolved versions of them--). It seems to be different again from various ways in which other contributors to these forums are using it. And both are different from the way the word is regularly used in publications on language studies.
Taken quite literally, your definition/description says that two words in related languages are cognate if they have a common source (good so far), and that this can be recognised by their having similar sounds and meanings in the two languages (not so good). If I proceed on that basis, several flaws emerge. Here are five types of case that your definition/description doesn't cover:
1. Similar sounds and meanings in the same language. Yard and orchard are both English words ostensibly derived from Old English geard. There is a resemblance in spelling and pronunciation, and also in meaning; and they are indeed cognates, but in the same rather than different languages. (The or- part of 'orchard' may be related to English wort, 'a plant', from which we also get root; so 'orchard' and 'root' may be cognate forms, both descended from 'wort'.) Some examples below are also in the category of "cognates within the same languages."
2. Similar sound, similar meaning, related languages, but not cognates. Such words are known as 'false cognates'--in this example, within the same language. English has a word yard meaning 'an area of land' (typically but not necessarily enclosed). It also has a word yard meaning 'a measure of area for land, equal to a quarter of an acre'. But these two words are not cognates; the first is (said to be) from Old English geard, a home (building and adjoining land), while the second is from Old English gerd, a straight branch or twig (used for measuring).
3. Similar meanings but different sounds, yet still cognate. My favourite example (again, within the same language) is 'beef' and 'cow'. The philologists can present a plausible (but unproved) case for their being cognate--both descended from a reconstructed earlier form that I can approximate as *gwous: one via Germanic *kwon, the other via Latin bovem.
4. Similar sounds but different meanings, yet still cognates. For example, gift in English means 'a present', but in German it means 'poison'. Clearly not cognates--but they are. Until quite recent times, the German word simply meant 'something given', as in English. It's only since the 16th century that its meaning has become restricted to 'something given in food and drink with the intention of killing'.
5. Different sounds, different meanings, but still cognates. English book and Greek phegos are a case in point, where phegos is a variety of tree called the 'edible oak'. The connection is that book comes from the fact that beech bark and tablets were used for writing runic inscriptions. Other related words are Latin fagus (beech tree) and Russian buzina (elder tree). (The alleged Indo-European word from which they are all descended is *bhagos. The beech, the elder, and the edible oak all produce edible fruit, relating them to the Greek verb phagein, 'to eat'.)
Now, your understanding of 'cognate' might encompass all of the above, but I'm afraid that that doesn't emerge from what you've written, and certainly not from how others have used the term. We are in Humpty-Dumpty Land, my friend, where each writer seems to use the same word in his or her own way, without consideration of how others use it or of its technical application. Confusion results on both sides--yours and mine.
The meaning of cognate in linguistics is this: Two linguistic forms (words, phrases, languages ...) are cognate if a plausible case can be presented for their parallel descent from a single earlier form, regardless of similarity or difference of sounds, spelling, or meaning, and regardless of whether they are in the same or different languages.
So why all this "ink"? "Rational discussion is impossible unless there is agreement on fundamental terms" (John von Neumann, approx.). Otherwise there's impenetrability! That's why I say!
|
|
| |
|
|
|
Donmillion
In: Acton, Middlesex
|
|
|
|
| But then nobody ever had to speak this NEW language...did they? |
Touché.
(I bleed).
So: Did they? How many people have managed to learn Latin, or any other "foreign" language, as a purely visual system, with no auditory component, even if it is not vocalised? I'm not saying it can't be done, but it must mean that no corresponding "sound" occurs in your head while you read or write it.
|
|
| |
|
|
|
|
|
|
|
|
|
Presumably, all literate deaf people - i.e. all deaf people in the West - which means it happens all the time.
|
|
| |
|
|
|
Ishmael
In: Toronto
|
|
|
|
Yes. It seems our conversation has been collectively headed in this direction; for I was thinking about this very matter last night as I lay in bed unable to sleep.
Is it possible that alphabetical language evolved absent vocalization?
The same human brain that assembles words and sentences from component sounds could presumably do the same using symbols, without the second system being a representation of the first.
The idea of assigning sounds to the symbols might have come about later as an innovation. Spoken language and alphabetical language might have originally been two separate systems that only later were seen to be combinable.
|
|
| |
|
|
|
Ishmael
In: Toronto
|
|
|
|
| Donmillion wrote: | | English has a word yard meaning 'an area of land'.... It also has a word yard meaning 'a measure of area for land, equal to a quarter of an acre'. But these two words are not cognates; the first is (said to be) from Old English geard, a home (building and adjoining land), while the second is from Old English gerd, a straight branch or twig (used for measuring). |
We do not recognize any methodology by which we can differentiate ancestors and offspring.
|
|
| |
|
|
|
|
The Library Index 
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in 
