Talk:English words of Greek origin

VfD

Survived: Wikipedia:Votes for deletion/List of English words of Greek origin Mikkalai 18:53, 4 Mar 2005 (UTC).

Old talk

50% of the words in the English language are of Greek origin, can we afford this page? seglea 20:56, 12 Aug 2004 (UTC)

Do you mean, can we afford a page that would be thousands and thousands of words long? It's a good point. I started the page because if we are going to have List of English words of Latin origin, then certainly we should have this one. But if you want to argue that we cannot afford either page, then I think you could easily make that case. -lethe ^talk

That is what was worrying me, yes. But I agree that in principle the page ought to exist. Maybe we could give some thought to ways of making it more economical, like listing types of words (with examples) rather than individual words. Or I suppose that once it gets too big (if we ever have the patience to put all those words in) it could be split into subpages by letter of the alphabet.

I'm also not sure whether I was right to put the more discursive stuff in this page - maybe it belongs elsewhere. But I think people need to be told why they should be interested in etymology.

seglea 00:05, 13 Aug 2004 (UTC)

Instead of sorting the subpages alphabetically, could we put the words into other categories? Ooh, alphabet, category, I should add those words.

Possible categories (some may overlap):
words ending in -sis (eg. analysis)
words relating to music (eg. music, orchestra, harmony, melody)
words relating to science
words relating to philosophy
words relating to poetry (e.g iambic, pentameter, dactyl)

A challenge: Whoever can think of a word to put in the 'W' list gets $23 from me.

Rather than trying to list English words derived from Greek, maybe it would be more useful to list Greek words that have given rise to English words. Again the number is theoretically unlimited, but avoiding the more obscure words we could surely come up with a list. For instance if it were not for possible copyright difficulties we could start with the list (about 20 pages long) in Bodmer's Loom of Language (1944).

I've now started such a list, based on the old "Greek Lexicon" article. It's at [[List of Greek words with English derivatives. rossb 09:56, 5 Jan 2005 (UTC)

Recognising Greek-derived words

I've deleted some of this. It previously implied that words containing x, or the digraphs æ or œ are generally from Greek - but in the actual text of the article (as opposed to the list of words) the majority of words containng x are Latin in origin, and the list itself seems to contain no examples of æ or œ. And I'm not sure what bayonet was doing.

Proportion of Greek words in English

An anonymous editor has just updated the proportion of Greek-derived English words from 10% to 25%. Can anyone provide a source for this?

These percentages are easily skewed. A lower percentage (lower than the 25 % stated above) is closer to reality. Alexander 007 01:16, 20 Feb 2005 (UTC)

The percentage is pretty funky. First of all, the vast majority of the 'borrowings' from Greek into English aren't really borrowings at all: they are _coinages_ using Greek morphemes, and never existed as Greek words (ancient or modern). Secondly, the vast majority of these words are technical and uncommon, so even if they represent 15% (or whatever) of the dictionary entries in an unabridged dictionary, they represent a far smaller proportion of the words used in texts. Finally, most of the words that _aren't_ modern coinages came through Latin, so a word can be counted as a borrowing both from Latin and from Greek, e.g. analogy < L analogia < Gk analogia.--Macrakis 21:08, 21 Feb 2005 (UTC)

The article Loanword quotes a more realistic (than the "25 %" figure quoted above) percentage from a 1970's study. According to that study, 5.32 % of English words are of Greek origin. Well, probably something like that. Alexander 007 05:05, 19 November 2005 (UTC)

Revealing an incredible statistical error
Greek does not account for 5,32% of the English language
20% would be a more accurate figure

I located the URL describing the survey on the origin of English words at: http://www.askoxford.com/asktheexperts/faq/aboutenglish/proportion?view=uk

A computerized survey of about 80,000 words in the old Shorter Oxford Dictionary (3rd edition) was published in Ordered Profusion by Thomas Finkenstaedt and Dieter Wolff (1973) which estimated the origin of English words as follows:

French, including Old French and early Anglo-French: 28.3%

Latin, including modern scientific and technical Latin: 28.24%

Old and Middle English, Old Norse, and Dutch: 25%

Greek: 5.32%

No etymology given: 4.03%

Derived from proper names: 3.28%

All other languages contributed less than 1%

This is a summary of the results of the above computerized survey

-	%	Entries
L	28,24%	22592
F	28,3%	22640
Gk	5,32%	4256
E & other	38,14%	30512
Total	100%	80000

In order to prove the validity of the above computerized survey, I performed a similar statistical analysis myself.

I used the Pocket Oxford Dictionary edition 1978. I took a sample of 13 random pages which included 327 entries and obtained the results below. Whenever there was an entry indicating F f. Gk (French from Greek) or L f. F f. Gk (Latin from French from Greek) I counted the entry as of Greek origin.

For example I counted the entry agony F f. Gk as of Greek origin.

I obtained the following results:

-	Entries	%
L	92	28,13%
F	31	7,95%
Gk	71	20,18%
E & other	133	40,67%
Total	327	100,00%

The results of Latin, English and Other, matched those of the survey. There was a striking difference however in the French and Greek results. I tried to figure out why. Using the same sample I performed the same statistical analysis but whenever an entry indicated " F f. Gk" I now counted it as of French origin. For example, I counted the entry agony F f. Gk as of French origin.

Then I obtained the following results:

-	Entries	%
L	92	28,13%
F	85	25,99%
Gk	17	5,20%
E & other	133	40,67%
Total	327	100,00%

Which is more or less the results of the computerized survey

This proves that 15% of the entries (12.000 entries out of the 80.000 total) appear as of French origin instead of Greek origin. Words like:

agony
air
agronomy
alabaster
allegory
amaranth
anachronism
anagram
apology
atrophy
austere
authentic
autocrat
autograph

to mention but a few, as well as 12.000 more entries is estimated to be included as of French origin, instead of Greek origin.

Unless someone can prove that the above list of 12.000 words are of French not of Greek origin, the above mentioned computerized survey should be considered as a fake as far as the Greek and French data is concerned.

It appears that a more realistic statistical distribution should be:

-	%
L	28,13%
F	7,95%
Gk	20,18%
E & other	40,67%
Total	100,00%

Until then, I propose to reconsider about the true statistical data.--Odysses 21:40, 22 November 2005 (UTC)

That's original research. We need published quotes. 20% seems rather high, and would need a citation in support of it. Even supposing 20 % would be accurate, it cannot be stated in the article based solely on your research. That is the policy.

Also, I'm not an expert on the percentages of the English lexicon, but it appears that you set Latin too high and French too low. Perhaps you included French words of Latin origin as being simply of Latin origin? That would be incorrect. For example, the English word anguish is from Old French anguisse, which evolved from Latin angustia ("narrow"), but it would be incorrect to include anguish as an English word from Latin (it is from French). Many other examples.

Q, "it appears that you set Latin too high and French too low."

A. Latin, English and Other categories are virtualy unchanged; French category is heavily changed. There is ample evidence that Greek words have been inputed as French words/entries.

My primary concern is not how many Greek words are in use today in English, but who originaly invented the Greek words.--Odysses 23:21, 22 November 2005 (UTC)

I'm still not sure how you trimmed down French so much. But yes, in the case of French words that derive from Greek they should be included as English words of Greek origin. But the ultimate origin of some of these words may trace to an ancient language other than Greek. Pegasus may be from an Anatolian language; tyrannos is of uncertain etymology and may be a loan from Asia Minor, etc. So if someone were seeking the "ultimate figure" based on who "invented" a given word, he may have to subtract some ancient Greek words of non-Greek origin. Alexander 007 23:31, 22 November 2005 (UTC)

I have no reason to object that ancient Greek language had also loanwords. If we compare modern English with ancient Greek this would be more apparent.

From the computerized survey, it appears in English that roughly 75% or 60.000 words are loanwords. I guess similar rates apply to French, German etc.

Let's examine a 1st century BC Greek Lexikon with 80.000 entries. We would also expect loanwords in this one too. It is natural to expect similar loan rates. However, from what we know today these loanwords are insignificant.

This is quite unusual and I really think we have a missing link somewhere.--Odysses 19:28, 24 November 2005 (UTC)

German is certainly much lower than 75%, possibly under 25%. This is due largely to their continued practice of compounding native (i.e. Germanic) word roots to name new things/ideas (whereas English turned largely to borrowing). French is much harder to determine mainly because if you consider the Latin language as the native, then words borrowed from Germanic, Gaulish, other Celtic languages, English, etc. are foreign. The problem here is that some of these 'foreign' loanwords were in use (proving that can be a problem) before Latin/Romance/French supplanted the native languages in Gaul. In the case of Greek, I think we should include loanwords (like angel and governor) in the Greek list unless someone can dig up the word (or the modern reflex of such a word) which these Greek words come from. Even if they can, the evidence will probably be speculative at best and the word should stay on the Greek list with a note telling of its possible history.--Hraefen 21:41, 24 November 2005 (UTC)

One example is the greek word titlos (in english it's title) which is (also) used in New Testament and derives from the latin titulus. +MATIA ☎ 21:50, 24 November 2005 (UTC)

Άγγελος is an ancient greek word (just checked it in a dictionary, it is also listed at List of Greek words with English derivatives). I don't know about governor. +MATIA ☎ 22:09, 24 November 2005 (UTC)

• Webster's 4th college ed. says govern is < Gr 'kybernan' < prob non-IE source and with angelos it gives only <?. They only do the <? when they think it stems back further than what there is hard evidence for. Anyway, I only brought these words up to point out that unless you have a good source for an etymology PRIOR to Greek (such as words that come to Greek from Hebrew/ Aramaic/ Semitic which are generally easy to trace) then those words should be included on the Greek list along with a note that the word may stem ultimately from some other source.--Hraefen 22:57, 24 November 2005 (UTC)

alternatively:

just an impulse, governor would be relevant to cūvern-or, or cyvern-or, cybern-etes, relevant to cybernetic--Odysses 23:11, 24 November 2005 (UTC)

I didn't knew about "govern is < Gr 'kybernan'" (and I really like that you use the < symbol, most editors didn't like it or didn't understand it when I used it in the past). I think 'cybern-' is obviously from the greek 'kybern-'. According to Yannis Koulakis (greek dictionary for etymology - ISBN 9602391359) άγγελος is an ancient greek word and "κυβερνώ < ancient κυβερνῶ, iapetic origin (my translation of ιαπετικής αρχής - I wonder what that means), compare with latin gubern-are" (note: the a of -are has also a ~ over it but I don't think I can type it). +MATIA ☎ 23:37, 24 November 2005 (UTC)

Iapetic an alternative to IE theory [1] --Odysses 18:33, 25 November 2005 (UTC)

...unless you have a good source for an etymology PRIOR to Greek (such as words that come to Greek from Hebrew/ Aramaic/ Semitic which are generally easy to trace) then those words should be included on the Greek list along with a note that the word may stem ultimately from some other source

Hraefen, could you please clarify on this?--Odysses 00:00, 25 November 2005 (UTC)

• What I meant was a word like messiah, even though it came to English through Greek, it is on and belongs on the Hebrew list. I may have misinterpreted what was being said above, but what I thought was being proposed was the removal of certain words from the Greek list which are thought to be or known to be words which the Greeks borrowed into their own language. If this IS the case, then my $.02 is that we should keep the loanwords on the Greek list unless someone has a good source to 'prove' it is a loanword. In such cases where it can be 'proven' I think those words belong on 1. the Greek list with a note telling of its inclusion on another list 2. that other list, be it Hebrew, Tukic, Arabic, etc. If the is NOT the case and no such removal was ever proposed then I apologize for wasting everyone's time.--Hraefen 19:47, 25 November 2005 (UTC)
  • In your example above, "messiah" is and should be on the Hebrew list, no question about this.
    Now, "astronaut" for example is classified as a word of Greek origin although it's a 20th century new word, since it was derived from άστρον + ναύτης (as per Argonaut).
    Likewise, it seems appropriate to me to classify all the words of Greek origin, i.e. words indicated in the dictionary as [Gr.] or [L f. Gr] or [F f. L f. Gr].
    Yet, this doesn't seem to be the case in the above computerized survey where there is evidence that 12.000 entries have been counted as of French origin instead of the older Greek origin.--Odysses 18:50, 28 November 2005 (UTC)

Missing loanwords in ancient Greek

This problem of missing loanwords in ancient Greek is not a new one.

In fact Plato deals in his entire dialogue Cratylus with those missing loanwords.

He attributes many important names and words of his time to ancient makers of the past.

...(like) in ancient times, we too seem to have said esia for ousia...

...and there, my friend, the modern interpreters of Homer may, I think, assist in explaining the view of the ancients. For most of these in their explanations of the poet, assert that he meant by Athene "mind" (nous) and "intelligence" (dianoia), and the maker of names appears to have had a singular notion...

http://classics.mit.edu/Plato/cratylus.html

There are even names that reflect a long history possibly several millennia ago.

Βόσπορος = βούς + πόρος (cattle passage) Today, no cattle can cross Bosporus, (without using the bridge anyway). Yet, several millennia ago it was probably a cattle passage. Probably a memory from the past.

Θεσσαλία = θέσις αλός (location of sea). It has been estimated today that the pass of Tempe, the canyon between Mt. Olympus and Mt. Ossa, has been opened towards the end of the glacial epoch, possibly 12.000 years BCE, when the ice melted, the sea level rose and water found a way out to the sea.

Μεσόγειος, L.. Mediterraneus (μέση, 'middle' + γη, earth) or Middle Earth. Is it normal to call a Sea... Earth? This could reflect memories of long ago, when during the glacial epoch, Mediterranean basin was possibly Middle Earth with many lakes.--Odysses 18:00, 25 November 2005 (UTC)

• I always took 'Mediterranean' to mean 'between the lands' (i.e. between Europe and Africa) and I would guess that the Greek Mesogeios was formed in a similar way--Hraefen 19:47, 25 November 2005 (UTC)

• • Good question, this would make more sense.
  • From my Gr ->En Dictionaries: μεσόγειος: inland, midland--Odysses 20:30, 25 November 2005 (UTC)

---

The actual figure probably is higher than 5.32 % which seems too low (maybe it's even 15%), which is why the article uses 5.32 % as a low-end figure (at least 5.32% of the English language). It's probably true that they did not count words like Tyrant, which though from French is in turn from Greek turannos, etc. The thing to do here is to find a high-end figure quoted (as much as...) in a reliable published source, but original research is not usable. Alexander 007 21:53, 22 November 2005 (UTC)

I think it would be nice to get a source that would show for example 20% from French, and of them 4% F f Gk 6% F f Latin. See also It's greek to me, from google. +MATIA ☎ 22:59, 22 November 2005 (UTC)

• To further complicate the issue...I think there should be some attempt made at differentiating between spoken/common/everyday words and scientific words. The inclusio or exclusion of words which a) are in a medical dictionary but b) not in a 'regular' dictionary will greatly affect the figure arrived at. i don't expect anyone to find a piece of hard, usable evidence for this but i think it should be addressed in the article somewhere.--Hraefen 23:13, 22 November 2005 (UTC)

• I agree with Hraefen. It is not fair to include tech and med words. However, it is very complicated to rate the "usefullness" of these words in everyday language. For example, judge for yourselves how frequently we use these tech and med terms: cell, anatomy, gynecologist, stomach, cancer, oxygen, meter, kilo etc. I therefore suggest that we include all technical and medical terms, since their degree of everyday use is subjective.

• Although I am not a fan of Mr.Portokalos in My Big Fat Greek Wedding who emphatically argued that every word comes from Greek, please, let us not deviate from the main subject, which is: Greek contribution to the English language is heavily undervalued with the 5.32% in the beginning of the article.

• User Odysses made a very impressive and verifiable presentation. This presentation was disputed by another user who did not sign his/her remark. The reason for the dispute was that "Even supposing 20 % would be accurate, it cannot be stated in the article based solely on your research. That is the policy." My view is that a research conducted by someone which can easily be repeated by anybody is self-proven and can be considered true.

• User +MATIA included a very interesting link (It's greek to me), from google. To this kind of links can be added the following: prof.Helena Dettmer of Iowa University states that 70% of English comes from Greek and Latin [2]. There are many more sites that you are invited to explore in order to further support this arguement. The article also needs the opinion of experts (linguists-philologists etc).

I suggest we adopt Odysses's research and change the article accordingly.

-NikoSilver 23:45, 29 January 2006 (UTC)

"More than 10%" Published by a reliable source.

The National Virtual Translation Center has published in its article regarding Greek Language that:

It has been estimated that over 10% of English vocabulary is of Greek origin.

The article can be found here: [3]

I am changing the openning sentence of the article accordingly and will continue to search for more complete evidence to support User:Odysses research (see above).

NikoSilver 19:57, 8 February 2006 (UTC)

I told you I'd continue searching.

The following links raise us to 12%:[[4]] and [[5]]

Changing the article again...

--NikoSilver 21:05, 8 February 2006 (UTC)

Reliable sources?

I don't think I'd accept the sources named above as reliable. One of them is obviously a Wikipedia mirror itself, and the others are apparently based on hearsay and/or don't say where they got their figures from.

There's specialised linguistic literature on this sort of topic, we really need to stick to that. I'd strongly suggest to only quote works that report on clearly identified empirical studies, whose samples and counting criteria can be verified. There's enough stuff of that kind around, so there's really no need in this case to stick to third-hand speculation.

For the moment, first thing I have here on my shelf: dtv-Atlas Englische Sprache (in German, but written by prominent linguists and quite reliable.) It gives a comparative table of three similar studies. One of them is the one that gives the 5.32% figure. The others are lower. - I'll need to dig into the small print to find out what the original studies were.

Anyway, we got to make three points quite clear:

1. Whether or not we count indirect loans (such as Gr > Lt > Fr > Eng) as belonging to the immediate donor or the ultimate donor, and what to do if Gr is itself only an intermediate donor (like in Hebrew > Gr > Lt > Eng), that's a thing we need to leave to the linguists to decide. If we report on a study, we can report on what their criteria were, but if they chose one set of criteria for their studies, then we're stuck with that and just have to report it. If there are different studies with different criteria, fine. I'd guess that most studies are likely to use the "only immediate donor" criterion (i.e. count Gr>Fr>E as "French"), because that's the only criterion that gives objective results in the cases of long borrowing chains where early members (pre-Gr, for instance) are not well known.
2. We should make a clear distinction in the article between type frequency (numer of Gr words in the dictionary) and token frequency (number of Greek word occurrences in running text). Since most Greek words are rare in text, token (text) frequency will be much lower than type frequency.
3. For token (text) frequency, we need to expect big differences between texts of different registers. Everyday conversation will have much lower percentages than academic prose, for instance. There are such studies, I could dig them out. Don't rely on random web searches please. I've got access to a English linguistics research library, and I'll be there on Tuesday.
4. For type (dictionary) frequency, we need to expect big differences between larger and smaller dictionaries. The proportion of Greek words rises dramatically towards the more esoteric areas of vocabulary; thus, a small dictionary that only lists basic vocabulary will have a much lower proportion of Greek words. This is the effect you see in the table from the book I mentioned:

Source language	Shorter Oxford E. Dict. (1964) 80.096 words		Advanced Learners Dict. of Curr. E. (1963) 27,241 words		M. West, A general service list of English words (1953) 3,984 words
	N	%	N	%	N	%
Romance	24,216	30.23	10,213	37.49	1,522	38.20
Latin	22,638	28.29	6,008	22.05	382	9.59
Greek	4.262	5.32	433	1.59	10	0.25
Germanic	21,047	26.28	8,670	31.83	2,028	50.89
Celtic	274	0.34	58	0.25	-	-

Lukas ^(T.|@) 21:32, 12 February 2006 (UTC)

Please find the CORRECT percentage

Reliable Sources

The National Virtual Translation Center where I got the 10%, is a US governmental service that is linked (among others) to: CIA, FBI, NSA, Homeland Security, Dept. of State and more. I think this source falls neither into the "wikipedia-mirror" category nor into the "hearsay" category.

The other two sites, where I got the 12%, are Answers.com and the Ministry of Foreign Affairs of the Hellenic Republic. Both of these sites are not a mirror of wikipedia (or they would state 5.32% which was the original text in the article). That, however, does not exclude them from the "hearsay" category. Therefore, I agree with Lukas that we need a more scientific approach.

Sorry, you were right, the relevant passage on the answers.com site was not Wikipedia. It's still third hand, though. The article does cite three works as sources, but it doesn't say which of them that figure is taken from. Lukas ^(T.|@) 06:16, 13 February 2006 (UTC)

Found it! The Columbia Encyclopedia, Sixth Edition. 2001-05 NikoSilver 17:51, 13 February 2006 (UTC)

Also, to be NPOV, the article has to mention ALL ways of estimating the percentage. More specifically:

Indirect Loans

Indirect loans can be attributed to either the FIRST (original donor) language that contributed the word, or to the LAST (immediate donor). The in-between donors have no real logical criterion to be included, apart from the fact that they enormously complicate things. Therefore, I propose we mention TWO percentages:

• Greek contribution to English as FIRST (original donor) language and
• as LAST (immediate donor) (I guess that's where the 5.32% comes from)

The 'original donor' category is problematic. Let's take, for example, "chrysobull". The first element, χρυσός, is a loan into Greek from the Semitic ḥārūṣ, and the second element is Latin bulla. On the other hand, the compound is clearly Greek. So do you want to count chrysobull as Greek because you can find that word in medieval Greek texts but not in Phoenician or Latin texts -- but how about aurea bulla? did that come first, and is chrysobull a loan-translation? I don't know... If chrysobull is Greek because the compound is first created in Greek, then how about "photography"? That compound never existed in ancient Greek; and we have a very precise origin for it -- Sir John Herschel coined it in 1839, and it was subsequently borrowed into French, Italian, etc. and eventually (1871) into Modern Greek. As for that matter, most of the Greek word-stock is Indo-European.... --Macrakis 16:07, 13 February 2006 (UTC)

It is problematic only for compounds, or have you found another example? If not, then compounds are a small percentage. I would think that compounds of different languages are even rarer. In these few cases, I would suggest that we look for both elements of the word. In your example of "chrysobull", that would be chryso (from Semitic) and bull (from Latin), NOT Greek, despite the fact that Greeks brought those two words together. So, that word would count 1/2 Semitic and 1/2 Latin.

Your "photography" issue is therefore clarified: photo and graph both come from Greek, despite the fact that a ...Sir put those two words together. So, that word would count 1/2 + 1/2 = 1 Greek. (win-some, lose-some)

I didn't understand the Indo-European comment. Indo-European language and Proto-Indo-European language are artificial, hypothetical and fictional. They may have never existed. They are just a logical explanation of how things may have happenned. Furthermore, if we go along with that, then we will find that EVERY word, of EVERY vocabulary has roots in the First Human Tribe that ever existed (before even the hypothetical Indo-European), which is a pointless conclusion. NikoSilver 18:36, 13 February 2006 (UTC)

That's exactly my point. The "ultimate" origin is not very useful. Clearly Herschel coined the English word photography using two Greek components, "photo-" and "-graphy". And whoever coined "chrysobull" as a new (medieval) Greek word coined it from two components which at the time of the coinage were already Greek words. I'm not sure what the point of considering "chrysobull" as a Semitic/Latin compound is. Conversely, the modern Greek word σινεμά clearly comes a French word whose full form, cinématographe, was coined in 1892. It would be a bit perverse, don't you think, to count σινεμά as a Greek word of Greek origin, as though it had been coined by Plato and transmitted through the manuscript tradition? --Macrakis 22:07, 13 February 2006 (UTC)

Dear Macrakis, please stop dealing with unimportant exceptions. We have an elephant infront of us and you are dealing with a hairline from its tail. We will never finish if we start with minuscule exceptions. Now if some French guy took two Greek words (kinisis and graph-) and put them together, later than Plato, we will count the compound Greek, because this is the definition of how we are going to do the research. To the point, it is logical to count the word as Greek (because he could have named it mobilo-scipta or whatever, BUT HE DIDN'T). This is covered in the Original Donor percentage. It would be logical too, to count the word French because that's where the two words were first put together. This is covered in the Immediate Donor percentage. It would be also logical to count them as Indo-European because that's where the hypothesis is that everything came from. WE DEFINE, though, in our research, that for compounds we will use the historically proven original origins of BOTH words. As you saw with "chrysobull" we may lose some Greek words in the process, while we may gain some others like "cinematographe" (only in the case of Original Donor percentage - in the Immediate Donor percentage it will count as FRENCH). NikoSilver 15:39, 14 February 2006 (UTC)

However, this is not the EXTREME case. It regards a SMALL percentage of the vocabulary, plus, in some cases it works against the Greek percentage. The EXTREME case (which I personally do not support) would be to call words of Greek origin, ALL those words that EVER passed through Greek (no matter if they were originally Semitic or Arabic or Latin). I am sure that this would give a great push to the Greek percentage, wouldn't it? Well, that is what the French contribution of 28.3% is about RIGHT NOW, since it just happenned that it is the LAST donor. Isn't this "unfair"? NikoSilver 15:27, 14 February 2006 (UTC)

By the way, though I certainly agree that Indo-European is not directly attested, it is a quite solid, well-established hypothesis. "Artificial" and "fictional" are peculiar descriptions of it. --Macrakis 22:07, 13 February 2006 (UTC)

For the "peculiar" description of Indo-European as "fictional" you are right, despite the fact that it is a solid hypothesis. You must admit, though, that "Indo-European" is an artificial name, that we gave to this hypothetical language. NikoSilver 15:27, 14 February 2006 (UTC)

Word Frequency of Texts or Oral Vocabulary

The English Vocabulary is defined as the TOTAL set of English words that exist rather than just the frequently used ones. However, since there are studies that actually show the frequency of use of any word (both in writing and in oral speech) I suggest we include these findings as well. Therefore, I propose to use another THREE percentages:

• Total Greek contribution to English, regardles of frequency of use,
• Text-Weighted: Greek contribution to English TEXTS, weighted by frequency of use and
• Oral-Weighted: Greek contribution to English everyday ORAL communication, weighted by frequency of use.

Size of Dictionary

This issue is irrelevant. We are interested in the percentage of Greek words in the English vocabulary, NOT in anyone's dictionary, no matter how big or small it is.

Not quite irrelevant, because the size of the English vocabulary is basically open-ended. The Greek vocabulary, more than anything else, has the property of being very easily extensible. You can always form new words through derivations. Does every technical term that any scientist used once in a research paper count? So, while in theory we might think of the "vocabulary" as an objective total, in practice any empirical study will have to decide on what sample of words to use, and that will always be some bigger or smaller dictionary. Lukas ^(T.|@) 06:13, 13 February 2006 (UTC)

Lucas, I think that one is covered by the weighted percentage of frequency of use. If you find the word "and" a billion times and the word "chrysobull" 5 times, then the second one will count 5 billionths compared to the first. Same with your scientist and his one word once. NikoSilver 18:42, 13 February 2006 (UTC)

Yes, if you are counting tokens, this is all true. But if you are counting types, the size of the dictionary does matter. Words like myriare, nanostere, and monochloroisocyanuric will be included if your dictionary is large enough. Does that tell you anything useful about the composition of English vocabulary? --Macrakis 22:28, 13 February 2006 (UTC)

Again, both versions of truth are covered. The 80.000 words or so Oxford Dictionary is a good approximation of the TOTAL English vocabulary, as perceived by most people. Also, frequency of use in Speech or Texts will essentially cancel out words like monochloroisocyanuric. NikoSilver 15:44, 14 February 2006 (UTC)

As a tendency, this goes in the right direction, but I can assure you these things have mathematical properties that make the actual statistics far more complex and fraught with far more uncertainties than you probably imagine :-) That's why I'd still urge we should not try to do our own estimations or rely on second-hand estimates that don't disclose how they did it, but wait till we find actual empirical studies in the original. Even though, as I said below, I must admit I'm quite disappointed I didn't find any at first sight. Lukas ^(T.|@) 22:56, 15 February 2006 (UTC)

I agree about the statistical complexity problem. Sometimes, in these cases, you just have to go word-by-word. We are only talking about a population of 80.000. Any college Greek class could do that in a week. Don't you think that it is ironic that there appear to be so many classical/Greek enthousiasts yet we still DON'T HAVE THE F****** PERCENTAGE!! NikoSilver 00:24, 16 February 2006 (UTC)

The size of the dictionary is ultimately determinative. The OED has more than 80,000 words, and has an unusually high percentage of Greek words: scientific terms and Victorian nonce words. I find a claim of 10% incredibly high; for example, although I write bookishly, and have not watched what I wrote, there is not a single only one Greek word in this paragraph. Septentrionalis 19:26, 16 February 2006 (UTC)

Dear friend, please read just 3 paragraphs above your edit, where it starts with the word "Again,". NikoSilver 10:38, 17 February 2006 (UTC)

And by the way, you could have written:

The mass of the lexicon is critically governing. The OED has more than eight myriads of lemmata, and has an atypical gigantic class of Greek lemmata: technical idioms and Victorian nonce words. I am skeptic about a claim of 10% as fantastically escalated. As a paradigm: although I write academically, and have not supervised what I wrote, there is not one only one twenty eight Greek words in this paragraph.

THAT would be bookish. NikoSilver 16:50, 19 February 2006 (UTC)

Summary

Things may look very complicated, but they actually aren't. We just need to fill-in the following table:

	Original Donor	Immediate Donor
Total Greek contribution to English vocabulary, regardles of frequency of use	??.??%	??.??%
Text-Weighted: Greek contribution to English TEXTS, weighted by frequency of use	??,??%	??,??%
Oral-Weighted: Greek contribution to English everyday ORAL communication, weighted by frequency of use	??,??%	??,??%

I think this pretty-much covers everything. I would therefore like to request the help of Lukas, as well as everybody else's who has the knowledge, background and access to such studies, in order to close the issue and make this article complete, accurate and, above all, NEUTRAL. NikoSilver 00:47, 13 February 2006 (UTC)

Literature search

I had promised I'd do a quick library search the other day. Unfortunately I didn't come up with as much as I had hoped. I have dug out the ultimate source of that 5.3% figure and some more info on how it was done (it's indeed only the immediate donor languages, as we figured out). I couldn't find anything about the text frequencies, which is a disappointment. As for statistics of indirect donor languages, not much seems to have been done at all. There is a reason for this, I guess: People who do these kinds of studies are interested in finding out something about English. The existence of a loanword always tells us something about the cultural relation between the receiving and the immediate donor language. If English speakers in the 14th century started saying "government" instead of "rīc", that tells us a lot about the cultural role of French at the time. It tells us precious little about Greek ("kybern-", if that is indeed the etymon). The English at the time were using the word because it sounded cool, because it was French. They didn't care whether it had been Greek a thousand years earlier. Maybe that's one of the reasons why linguists haven't felt such statistics were much worth it. - I'll be back with the bibliography details later. Lukas ^(T.|@) 22:48, 15 February 2006 (UTC)

My friend, I was disappointed too, when I searched the web before that. Sure, the influence plays a role, but there are other important things too. I think these stats ARE WORTH it. After all, we are talking about one of the greatest cultural contributions that ever existed. Why not also in vocabulary? NikoSilver 00:18, 16 February 2006 (UTC)

Proposal (one more)

How do you feel about the following paragraph:

Greek language has contributed directly (as an Immediate Donor) 5.32% of the words of the English vocabulary. However, XX.XX% of the English words have one way or the other passed through the Greek language. This percentage, includes both words that are of Greek origin, yet ended indirectly to English (mainly through French and Latin) and words that had been imported by other languages to Greek before they were eventually imported to English. It also includes all compounds that have a Greek root in any part. Practically, however, a large number of these words are seldomly used in everyday speech and texts.

I am sure that this simplification solves thousands of statistical/mathematical research problems, plus it gives us the maximum. Having 5.32% as the minimum in the article, I suspect that showing also the other extreme gives us a good NPOV. After all, the article is called English words of Greek origin, NOT English words of immediate Greek contribution. Can anyone help find what this XX.XX% actually is? Any opinions? NikoSilver 00:18, 16 February 2006 (UTC)

I think the paragraph would be a good start - if we could actually find a reliable figure for "XX.XX%". More later. Lukas ^(T.|@) 08:14, 16 February 2006 (UTC)

Thanks Lukas! We are expecting more opinions on this matter for the next few days. After that:

Everybody: We are trying to locate the percentage of English words that were imported directly or indirectly from Greek. NikoSilver 14:19, 16 February 2006 (UTC)

---

I haven't looked at this page for over a month and I was amazed to see that the interesting discussion goes on.

First I would like to thank Lukas and NikoSilver for confirming what I proposed earlier. Some 12.000 words of Greek origin have been counted French origin as the immediate donor. Yet the same does not seem to apply to Latin from French. I wonder why?

Similarly, if an English Museum would buy an ancient Greek artifact from a French archeologist or antiquarian and made a few rectifications or modifications to it, would it be justifiable to classify the artifact under French antiquities?

Gongratulations! This is a Giga-Metaphor!NikoSilver 22:07, 19 February 2006 (UTC)

Back in the 14th century very few would know or care if the words "agony", "allegory" or "government" is French or Greek, but it makes a lot of difference today in the 21st century.

Macrakis, you don't seem to understand that in my analysis I have simply counted what was in the Pocket Oxford Dictionary edition 1978. At this stage, I assumed their etymology was correct. Therefore I cannot see your point on cinematograph and chrysobull. Besides, there is no such entry "chrysobull" in the above dictionary.

As far as the "Size of the dictionary" is concerned, perhaps this is relevant:

...vocabulary researchers reported figures ranging from 2,500 to 26,000 words in the vocabularies of typical grade 1 students and from 19,000 to 200,000 words for college graduate students (Beck & McKeown, 1991). [6]

Based on the above, I would think that anything between 80.000 to 200.000 words would be valid.

Do you mean distinct words, or words with all conjugations/inflections? NikoSilver 22:07, 19 February 2006 (UTC)

As for the percentage of English words of Greek origin, I haven't located anything worth mentioning, although I am dealing with this for years. Perhaps we could create such a list ourselves, here in Wilipedia. Even with a few hundred or thousand entries we could estimate statistically the proportion. --Odysses 20:19, 19 February 2006 (UTC)

Sorry for editing inside your text, makes more sense though. Glad you are back! NikoSilver 22:07, 19 February 2006 (UTC)

Suggestion for anyone who wants to edit his opinion here

Please read the whole discussion, starting at least from the heading:

"More than 10%" Published by a reliable source.

Otherwise, your comments may have already been covered. NikoSilver 15:59, 14 February 2006 (UTC)

An approximate estimation of English words of Greek origin

In the List of Greek words with English derivatives there are well over 300 entries in the letter "Alpha" alone and the list is far from complete yet.

Some of these entries contain quite a few derivatives. Other entries may contain only one or two derivatives.

If we could estimate the number of English derivatives per entry on average, we might be able to get a reliable figure.

Let's examine the roots auto- aero- allo- and anthropo-, located in the "The American Heritage® Dictionary of the English Language, Fourth Edition"

No	auto-	aero-	allo-	anthropo-
URL	[7]	[8]	[9]	[10]
1	automatic	aeroballistics	allocholesterol	anthropology
2	autoantigen	aerobatics	allochton	anthropomorphism
3	autobiography	aerobic	allodium	anthropopathy
4	autocatalysis	aerobicize	allodontidae	anthropometrics
5	autochthon	aerobics	allogenes	anthropomancy
6	autochthonous	aerobiology	allograft	anthropopath
7	autoclave	aerobiosis	allography	anthropoglot
8	autocoid	aerobium	allolactose	anthroponyms
9	autocracy	aerodrome	allometry	anthroposophical
10	autocrat	aerodynamic	allomorph	anthroponym
11	autodidact	aerodynamics	allopath	anthropophilic
12	autodyne	aerodyne	allopatry	anthropomorphists
13	autoecious	aeroembolism	allophony	anthropocentrism
14	autoerotism	aerogram	allopoiesis	anthropomorphics
15	autoerotic	aerology	allopurinol	anthropoidea
16	autogamy	aeromagnetics	allosaur	anthroposemiotics
17	autogenesis	aeromechanics	allotroph	anthropophobia
18	autogeny	aerometeorograph	allotropic	-
19	autogenous	aerometer	allotropy	-
20	autogyro	aeronaut	allergy	-
21	autograft	aeronautic	-	-
22	autograph	aeronautics	-	-
23	autography	aeroneurosis	-	-
24	Autoharp	aeronomy	-	-
25	autohypnosis	aeropause	-	-
26	autointoxication	aerophagia	-	-
27	autoloading	aerophobia	-	-
28	autologous	aerophore	-	-
29	autolysate	aerophyte	-	-
30	autolysin	aeroplane	-	-
31	autolysis	aeroponics	-	-
32	autolyze	aerostat	-	-
33	automata	aerostatics	-	-
34	automaton	aerotaxis	-	-
35	automate	aerothermodynamics	-	-
36	automated	-	-	-
37	automatic	-	-	-
38	automation	-	-	-
39	automatism	-	-	-
40	automatization	-	-	-
41	automatize	-	-	-
42	autonomic	-	-	-
43	autonomy	-	-	-
45	autonomous	-	-	-
46	autonym	-	-	-
47	autophagy	-	-	-
48	autoplasty	-	-	-
49	autopolyploid [11]	-	-	-
50	autopsy	-	-	-
51	autoradiogram	-	-	-
52	autoradiograph	-	-	-
53	autosome	-	-	-
54	autotomy	-	-	-
55	autotoxemia	-	-	-
56	autotoxin	-	-	-
57	autotroph	-	-	-

Similarly there are numerous other derivatives in roots like, -agogy anti- -cracy -cycle geo- -hex- -logi -logy mega- mono- neo- -nomy -oct- -oid -onym -pent- -philia -phobia -scopy -trophy -urgy etc. [12]

Assuming an average of 10 English derivatives per entry and 300 entries for the letter Alpha,

10 x 300 = 3,000

we get 3000 English derivatives for the letter Alpha.

Since letters like i.e. omega (Ω) and xi (Ξ) have very few entries, I have ignored them, so multiply by 20 to obtain a total for the whole alphabet:

20 x 3,000 = 60,000 English derivatives

We also need to know the number of words in the English language: The following URL [13] gives a total of 230,000 words, including obsolete words.

Based on the above analysis, the required proportion would be:

60,000
-------- = 26 %
230,000

I believe that the above rate 26 % is a well documented, reasonable and realistic figure.--Odysses 13:21, 21 February 2006 (UTC)

Comments

This is all very nice :-) But still, I don't think it works. First, it's original research again, of course. Then, the estimate of entries per initial letter is insecure. Some letters have very productive prefixes (like "allo-", "auto-"), others haven't. Third, the proportion 60,000 / 230,000 might be comparing apples with oranges. Who knows how the "230,000" were arrived at? Are words like "autopolyploid" actually included in those "230,000"? If you want to have a serious comparison, you'll need to count Greek words from an actual dictionary, and compare that number with the number of words in that dictionary. And by the way, if you count "autopolyploid", what about "autopolyploidity", "autopolyploidism", "autopolyploideering", "autopolyploidous", "autopolyploidification", "autopolyploidificant", "autopolyploidificationalist", "autopolyploidificationalism", "anti-autopolyploidificationalism", "anti-autopolyploidificationalistically", and "anti-anti-autopolyploidificiousnesses"? Lukas ^(T.|@) 14:47, 21 February 2006 (UTC)

Thank you for your comments Lukas,

Good point, I forgot to mention that I located these derivatives in the "The American Heritage® Dictionary of the English Language, Fourth Edition" [14]. Every single entry from the above table is from this dictionary. I estimate it contains somewhere around 200.000 entries.

I have now added the URL in second row of the table.

There is no entry however for ... "autopolyploideering" [15] or "autopolyploidity" [16] etc. Sorry but I cannot add them to the above list :-)

I will also try to locate the number of enties in this dictionary.

I found something in [17]

The American Heritage® Dictionary of the English Language. Over 90,000 entries feature 10,000 new words and senses, 70,000 audio word pronunciations, 900 full-page color illustrations, language notes and word-root appendixes.

I am not sure if this is correct.

Dividing 60,000 / 90,000 = 66,666 %

I will reject it. Too bad.

Borrowing and coinages

It seems to me that there are two distinct categories of words which you might say are "of Greek origin" in English: words borrowed from Greek, and words coined using Greek roots.

The borrowed words existed at some point in the Greek language (regardless of their further etymology) and came into English mostly indirectly, via Latin, French, etc.. These include cases like 'bishop' < επίσκοπος as well as the more obvious ones like 'philosophy', 'theater', 'agony', 'chrysobull', and even 'moussaka' (originally Arabic, of course).

The coined words were created in the European languages using Greek roots. These are sometimes lumped into the category of "New Latin" in dictionary etymologies, but of course it would be more accurate to call it "New Greek" in this case.

Then there are a few cases like 'chemistry' which has a surprisingly convoluted and interesting history. (Look it up!)

I don't think it makes sense to lump genuine borrowings from Greek together with coinages which never existed in Greek (unless they were borrowed into Modern Greek). --Macrakis 15:06, 21 February 2006 (UTC)

I agree with this. -lethe ^{talk +} 16:36, 21 February 2006 (UTC)

We need not agree or disagree with this. What we need to do is to provide both versions. As I wrote before, words like cinematograph (kinisis+grapho) could have been named mobiloscripta or something but they WERE NOT. So, there are 2 logical approaches (each correct) for the origin of such words:

1.French, because a French person put those two Greek words together for the first time and

2.Greek, because the French person could have taken any 2 words from any other language, yet he chose two Greek ones.

In such cases, both percentages should be taken into account. So we will simply present them both.NikoSilver 16:54, 21 February 2006 (UTC)

---

By Greek we all mean Ancient Greek language of course.

Here we compare Ancient Greek versus Modern English. Modern Greek has nothing to do with this. And frankly it never occurred to me to ask any copyrights for English words of Greek origin used today.

In fact through English, Modern Greek acquires some Ancient Greek words, like αεροδρόμιο (aerodrome) [18] or μάους (mouse in PCs) (from Old English mūs) [19] (from Gk mūs) [20]

Let's for the shake of argument assume that the Ancient Greek word κλών (klōn) [21] never existed.

I cannot see how a "coined" word cloning could then exist today in the European languages. Cloned sheep Dolly would, but it would be called something else, like twigged or branched.

Similarly if we assume that the Ancient Greek words άστρον, ναύτης, and κόσμος never existed, how could we ever have today the "coined" words astronaut and cosmonaut.

By the way, most dictionaries consider the word astronaut of Greek origin [22] even though no astronauts and cosmonauts existed then?

I wonder if a French or German dictionary lists the word astronaut as of En. or Gk origin.

I am sorry, I dont think that borrowed or coined makes a lot of difference.--Odysses 17:43, 21 February 2006 (UTC)

Comment #2

What Odysses did in his second study "An approximate estimation of English words of Greek origin" above is very analytical for part of the letter Alpha and could be generalised, up to an extent, as an approximation, for the rest of the dictionary. However, the hypotheses that have to be assumed in order to reach the conclusion, do not justify the end result as proven by a "Reliable Source".

In order for this to be done, I suspect that the same study should be conducted in the total set of words of the dictionary under examination.

We need a Reliable Source that justifies the end conclusion. Therefore, at this point, I would like to introduce to the rest of the editors of this article, a book that has been publihed by the Greek linguist Aristidis Konstantinidis and has already been discussed between me and Odysses in his talk page. The book is in Greek. You can find reviews and comments on this book here: [23], [24], [25], while the author's page is still under construction here: [26].

For those who cannot read Greek, the reviews say that the author proves that 1 out of 4 words of the English vocabulary are Greek, apart from Botanological, Chemical and Zoological terms. If we include these, the analogy shifts proportions: 3 out of 4 words are Greek. The reviews also mention that never before has there been any serious research on the matter and provide certain examples of other inconclusive and incomplete researches of the past. That may be the reason (my friend Lukas) that we cannot find any published article or research yet!.

We are expecting to receive more information on the methodology of the research in the book and assuming it is substantial, we will edit the article to 25%. NikoSilver 16:38, 21 February 2006 (UTC)

Yep, if somebody could actually get hold of that book, that would be nice, I'm sure we could find something to work in. However, the estimated 3/4 figure still looks a little bit endoprokatautopolyploidophoricalizing to me. Not to say supercallifragilistixpialidotious. Lukas ^(T.|@) 19:28, 21 February 2006 (UTC)

There is a list of some of the words counted as Greek origin in one of the blurbs cited above:

pause, bottle, channel, scenario, gas, bomb, baleen, calm, camera, cherry, police, pilot, church, chestnut, blame, turtle, scar, cosmetics, trophy, caravel, canopy, brilliant, austere, butter, copper, gulf, devil, hilarity, truck, caliber, alms, charivari, box, lard, buffalo, goblin, idiot, mandrel, squirrel, tiffany, naudea, turn, dyne, spermaceti, cretin, parole, scheme, pain, penalty, pirate, priest, clerk, crergy, bishop, monk, diplomat, paper, sketch, chart, deacon, dish, mass, talent, surgery

Now this, I find interesting. Much more interesting than "photography" and "megameter". Of course, many languages and cultures have contributed to the sense-developments of these words. --Macrakis 19:46, 21 February 2006 (UTC)

By the way, Aristidis Konstantinidis is characterized as a "Greek linguist" above, but the referenced sources don't mention his education or current affiliation, and his book is self-published. --Macrakis 19:49, 21 February 2006 (UTC)

You are right. He is not a linguist. He is a lexicographer. Maybe we should just at least click to see the back cover of the book, here: [27] and read some comments of linguists and professors who read the book.

Another good idea is to read the following part of the review by the third site which I provided (sorry I didn't get into the trouble of translating it):

Ολες αυτές οι λέξεις, που έχουν καταγραφεί, στην έρευνα του κ. Κωνσταντινίδη, είναι λέξεις που οι άγγλοι και οι αμερικανοί, στα εγκυρότερα λεξικά τους, αναγνωρίζουν και παραδέχονται, ότι έχουν ελληνική προέλευση.

Η έρευνα λοιπόν, δεν έχει βασιστεί σε προσωπικές ερμηνείες ετυμολογίας.

Επιπλέον πολλά λεξικά πλην αυτό της Οξφόρδης σταματούν δυστυχώς στη λατινική ρίζα και δεν αναφέρουν για διαφόρους λόγους, εάν προήλθε από την ελληνική η συγκεκριμένη λατινική ρίζα π.χ. η λέξη Electric, ηλεκτρισμός αναφέρεται ότι προέρχεται από το λατινικό electrum, δεν γίνεται όμως αναφορά ότι αυτό προήλθε από το ελληνι- κό ήλεκτρον, το κεχριμπάρι.

Το λεξικό της Οξφόρδης, περιλαμβάνει 10.500 ελληνικές λέξεις, που αντιστοιχεί στο 21,6% αυτού του λεξικού.

Λέξεις της αρχαίας ελληνικής, που ήταν δάνειο από την περσική, όπως η λέξη αγγαρεία ή εβραϊκές, όπως η λέξη σατανάς, δεν έχουν περιληφθεί στην εργασία.

A search in Google for his name, would provide you with the following links-reviews-critics: ΤΑ ΝΕΑ 1, ΤΑ ΝΕΑ 2, Publisher site.

NikoSilver 21:51, 21 February 2006 (UTC)

Uh-oh. It does seem he's an amateur. According to [28], he studied law, worked as an English translator, manager, and freelance management trainer. And if the following summaries (in poor German) on another advertisement website are representative of the book [29] it isn't exactly confidence-inspiring: "Eine Unterwanderung des Englischen durch das Lateinische begann, als die Römer 55 n.Chr. England einnahmen und bis 410 n.Chr. besetzt hielten. In dieser Periode der Verrömerung der Kelten etablierte sich langsam auch der griechische Wortschatz im anglosächsischen Sprachraum durch Begriffe wie Anemie, Aorte, kakoethes, emphasis, ainigma etc. Des weiteren änderte sich vieles durch die Lehre des Christentums. 597 verließ der Heilige Augustinos Rom und begab sich nach England um dort das Volk zu christianisieren. Da das Evangelium aber auf Griechisch verfasst war, wurden im Englischen unbekannte Wörter einfach adaptiert" (My translation: "The influence of Latin on English began when the Romans occupied England in 55 AD and kept it until 410 AD. During this time of Latinization of the Celts, Greek vocabulary slowly established itself in the Anglo-Saxon linguistic world, with words such as αναιμία, αόρτα, κακοηθής, έμφασις, αίνιγμα etc. Much more changed with the new doctrine of Christianity. In 597 St. Augustine left Rome for England in order to Christianize the people there. But since the gospels were written in Greek, words unknown in English were simply adapted.") Well, this is really a jumble of nonsense, sorry.

Well, we'll wait and see, I'd say.

Lukas ^(T.|@) 23:16, 21 February 2006 (UTC)

Please excuse my lack of education on the subject. Can you explain why it is a jumble of nonsense? NikoSilver 00:26, 22 February 2006 (UTC)

Well, (1) there were no English in England when the Romans were there, so they can't have started taking over vocabulary from them at that time (and certainly not "emphasis" and "ainigma"). And (2) the gospels St Augustine brought to England later were in Latin, not in Greek (that was the start of the first wave of Gr>Lat>Eng indirect loanwords, of course). And if I remember right, "baptism" wasn't even borrowed yet at that early time, they first invented an English equivalent, "fulltūm" or something like that. (Those Germanic barbarians weren't very quick to pick up foreign anyway. Some time around 800 AD an English king got so exasperated with the poor state of Latin knowledge among his clerics that he ordered to have whole books translated into English, so that at least a few people would be able to read them - which was a really novel idea at the time. :-) Lukas ^(T.|@) 00:57, 22 February 2006 (UTC))

Please bear my scepticism and my lack of knowledge on the matter, but your answer raises some more questions. Naturally, I agree, there were no English people as we know them today, but there were tribes collectively called today Brythons who merged after the Romans left (5th, 6th cent.) with other Germanic tribes (mainly Jutes, Saxons and Angles) to form today's English people. I find it very possible that these people's (the Brythons') vocabulary was influenced by their Roman conquerors (ofcourse we would only mean indirect loan words from Greek). Why would they have to wait for 550 years (for St.Augustine) before they got 1 word from the Romans? Slow-learning Barbarians or not, we are talking about some 20 generations!

Ah, there's a misunderstanding here: the Angles and Saxons were not other Germanic tribes, they were the first Germanic tribes on the island. The ancient Britons spoke, well, Brythonic. Which is the ancestor of Welsh, i.e. a Celtic language. And while it stands to reason that the Brythonic and the Anglosaxons mixed genetically, almost nothing of the Brythonic language (at least not vocabulary) was taken over by the English. So, even if they had some Gr>Lat>Bryth loanwords, there was no way to pass those on to English. Lukas ^(T.|@) 23:25, 22 February 2006 (UTC)

Then how can we explain Breton language words like skol(=En.school<L.<Gr.σχολή) like in the Examples Section of Breton Language here in wikipedia? How can we make hypotheses about the existance of a whole language (Proto-Indo-European) and not assume the obvious: That people under Roman occupation for 550 years would most probably import certain Latin/Greek words and later on when they "mix genetically" to pass those words to "Germanic barbarians" who probably didn't have any word for such civilised things like "school"? NikoSilver 11:43, 23 February 2006 (UTC)

Oh, of course Celtic languages also borrowed from Latin, at various times, before and after. But it's highly unlikely that the ancient Britons passed Latin vocabulary to the Anglo-Saxons, because they apparently didn't pass any vocabulary to them in the first place, be it native or non-native. There are hardly any Celtic loanwords in English from that time. It's quite unlikely that they should have passed on only the Latin/Greek stuff and not also a few native Celtic words along with them. Besides, if any Greek loanwords had passed through Celtic on their way, we'd in all likelihood be able to see traces of the Celtic influence in the actual forms of these words, and to the best of my knowledge such a thing has never been claimed. This would be indeed revolutionary. In the absence of the actual book, we may assume that all this is probably just an incompetent and misleading summary on that web site, but the fact that the advertiser could get the contents of the book that wrong doesn't speak too well of the book either, in my view. Lukas ^(T.|@) 11:53, 23 February 2006 (UTC)

What if the guy has actually found evidence of these word borrowings even so long ago? Or even, what if he is just making a hypothesis and the words (which ARE Greek) were simply borrowed later. Does that make the end conclusion (percentage of Greek words) false?

I read that he devoted his life in this. This is the second book he publishes on the matter (the first one was in 1991!). He was given the first award from the Athens Academy in 1994. He can't be that bad. His credentials on the matter sound better than any "educated amateur's" (sic).

We really don't need to start an "online lesson" (you being the teacher, ofcourse), since we don't even have the book yet. So let us do what you proposed: Wait and see... NikoSilver 23:11, 22 February 2006 (UTC)

---

I admit that etymology is not my science, but I find it fascinating.

It seems that there is a misunderstanding, which I will try to short out.

What you imply Lukas from your brief history account above, is that this article includes this type of entries:

Type of entry	Example
1. Gr>Lat>Eng	amnesia
2. Gr> Eng	aorist [30]

(Hence the estimated 5,20% or 12%)

But it doesn't include this type of entries:

Type of entry	Example
3. Gr>Lat>Fr>Eng	philosophy [31]
4. coined words or New Greek	astronaut [32]

(Hence the disputed "more than 12%")

Question 1: Is this correct that it includes only type 1 and 2?

Question 2: What should be the title of a list that would include all four types? (appart from supercallifragilistixpialidotious anyway :-)) --Odysses 12:06, 23 February 2006 (UTC)

Sources, at last

I still didn't make good on my promise to present the sources I had found. So, here goes:

The 5.3% figure comes from Scheler (1977: 48). If I understood it correctly (and I only had time to skim it, I must admit), it includes your types (2) and (4) above, but it notes that the line between (1) and (2) is often difficult to draw. Scheler looked at three different word lists: A medium-to-large-sized dictionary (80,096 words), a somewhat smaller dictionary (27,241 words), and a list of basic functional vocabulary (3984 words). In these three lists, the figures were 5.3%, 1.6%, and 0.25% respectively. The full statistics is the one I gave in the table above under #Reliable sources?. As this figure includes the coined words, the figure for the real, direct loans (category 2) is likely to be even lower.

In a comparable study (Roberts 1965), there was a small dictionary of 10,000 words, and the figure for Greek was hidden in a lump category of "other languages" with 2.3%, so Greek alone would probably range somewhere around 1%.

Now to Odysses' questions above:

If you were to work from an even larger dictionary including technical scientific vocabulary from all the different sciences, then the weight of category (4), the coined words, will rise disproportionally. Probably that's part of what Konstantinidis has done. I can well imagine that you can tweak the figures to more than 25% if you include the right branches of science (there are obviously some disciplines that produce an obscene amount of specialized vocabulary).

If you were to stay with the dictionary size Scheler had chosen (c.80,000 words of mainstream, not-too-specialized English), but include the indirect loans (your categories 1 and 3), then I guess a figure around 12% might be plausible. Lukas ^(T.|@) 12:40, 23 February 2006 (UTC)

So how do we explain User:Odysses study Revealing an incredible statistical error which was conducted on a random sample of an 80,000 words vocabulary (the Pocket Oxford Dictionary edition 1978)? NikoSilver 17:21, 23 February 2006 (UTC)

Hm, you're right, Odysses' figure of around 20% would agree pretty well with Konstantinidis figure, which is also in that range. I'm crap at mathematics, and I had probably first underestimated the statistical value of the size of Odysses' sample (327 entries) - but now that I look at it, I get a confidence interval out of Odysses' figures of the real value being somewhere between 18 and 25 percent, at confidence margins of 5%. Brrrrr, I'm sure Macrakis is more competent than I am at such calculations. I'd say, if we explain the thing about direct vs. indirect loans in the article, and that about the importance of the population size (smaller vs. larger dictionaries), and quote the Scheler and the Konstantinidis thing (in the absence of anything better), we could use the 5% and 20% figures. Lukas ^(T.|@) 18:19, 23 February 2006 (UTC)

If it's any help, the Computerized survey used about 80,000 words in the old Shorter Oxford Dictionary (3rd edition), and I used the Pocket Oxford Dictionary edition 1978 which is similar.

Lukas, I aggree with the 5% and 20% figures.--Odysses 19:37, 23 February 2006 (UTC)

References

Consulted

• Hughes, Geoffrey (1988): Words in time: A social history of the English vocabulary. Oxford: Blackwell.
• Roberts, A. H. (1965): A statistical linguistic analysis of American English. The Hague: Mouton.
• Scheler, Manfred (1977): Der englische Wortschatz ['English vocabulary']. Berlin: Schmidt.
• Stockwell, Robert, Donka Minkova (2001): English words: history and structure. Cambridge: Cambridge University Press.

Not consulted

• Gramley, Stephen (2001): The vocabulary of world English. London: Arnold.
• Lipka, Leonhard (1990): An outline of English lexicology: Lexical structure, word semantics, and word formation. Tübingen: Niemeyer.
• Smock, J. C. (1931): The Greek element in English words. London.
• Wermser, Richard (1976): Statistische Untersuchungen zur Entwicklung des englischen Wortschatzes ['Statistical studies of the development of English vocabulary']. Bern: Francke.

Lukas ^(T.|@) 12:40, 23 February 2006 (UTC)

These are surely interesting sources. Good job Lukas!

Clarification: In my early postings in November I assumed that this figure included all types (1) (2) (3) and (4). Following your explanation earlier however, I find it reasonable to include only direct loans, hence type (1) and (2).

It might be a good idea however to add a little note somewhere in the article explaining that the article does not include words of indirect origin, with one or two examples. This will save you from many lengthy enquiries by irritating editors like me :-)

Q1. Did you mean Scheler included types (1) and (2)?

Q2. Did you come across any Greek researcher(s) during your search?--Odysses 19:39, 23 February 2006 (UTC)

When will the article start growing?

Is there a chronology? It is worth mentioning that English was made official only in 1372(?), and the English language is the most recent modern language in the world. More important, it is the only western tongue that was not shaped directly through Greek (like the Latin, Armenian, Slavonic and Germanic tongues), but through German and French. It is a 'krama' of words and consists of two dialects: one where Latin and Greek words predominate, and one where the dominant vocabulary is Germanic. Loipon? (unsigned contribution by User:Politis)

I am not sure why it matters when English was made official, what it means for a language to be "most recent" or to be "shaped directly through Greek", and your usage of 'dialect' is not standard. But in any case, I don't see their relevance to this article. There is an article on History of the English language. --Macrakis 20:13, 23 February 2006 (UTC)

This is a very interesting discussion page, I was just wondering when it would translate into an Article. Thanks for the History of English link. But I suggest that this Article might receive a boost with a couple of paragraphs, outlining the basic chronology on the introduction of Greek words into a Germanic language it helped transform itself into English. From the arrival of Theodorus of Tarsus who opened a Greek and Latin academy in the 7thC, to the education of English students by Greeks scholars in Italy in late 15th and 16thC, the scholarship of the Bible, through Greek theologians, in 17thC, and the discovery of Classical Greek texts and plays. Politis 14:27, 24 February 2006 (UTC)

For once I agree with Macrakis. However, also, Politis has a point in the TITLE of his comment. In view of his anticipation and Lukas', Odysses' and my agreement in changing the article as first proposed by Lukas, I propose the following paragraph as the initial paragraph of the article:

---

The Greek language has contributed to the English vocabulary in four ways:

1. directly as an immediate donor,
2. indirectly through one intermediate language, as an original donor (mainly through Latin and French,
3. indirectly through more than one intermediate languages again as an original donor, and
4. with modern coinages or new Greek.

The size of the Greek word participation, in an average English dictionary of 80,000 words, which does not include specialised technical and scientific words, ranges from 5.32% to about 25%, based mainly on which of the four ways of contibution one chooses to define as of Greek origin. If all technical and scientific specialised words are also counted for, then this percentage increases dramatically.

---

Of course we should include at the end of the article the references above mentioned by Lukas (Scheler, Konstantinidis and if necessary others). Opinions? NikoSilver 23:40, 23 February 2006 (UTC)

Yep, something along those lines will be fine. I'd say that the distinction between "one intermediate" and "more than one intermediate" is a bit artificial and complicates matters more than necessary. I apologize for not making more contributions to the article directly right now. As for the things about "official" language since 1372 and "most recent modern language", please leave it out, both are rather inaccurate - and not really related to the understanding of the vocabulary issues. Lukas ^(T.|@) 08:14, 24 February 2006 (UTC)

Changed the article accordingly. NikoSilver 10:47, 24 February 2006 (UTC)

---

The Greek language has contributed to the English vocabulary in three ways:

1. Directly as an immediate donor,
2. Indirectly through other intermediate languages, as the original donor (mainly through Latin and French), and
3. Through modern coinages or new Greek.

The size of the Greek words in an average English dictionary of 80,000 words, which does not include technical, scientific and highly specialised words, ranges from 5% to 25%, depending on which of the three ways of contibution one chooses to define as of Greek origin. If all technical, scientific and highly specialised words are included, then this percentage increases dramatically.

Yep, looks good. I propose the above minor changes (shown in bolt).--Odysses 12:11, 24 February 2006 (UTC)

Also, perhaps no bolted words needed, only link words.--Odysses 12:58, 24 February 2006 (UTC)

No reason to disagree by me. NikoSilver 13:49, 24 February 2006 (UTC)

Chronology etc

Everybody, please check what Politis says just some paragraphs above at 14:27, 24 February 2006 (UTC), cause it may have passed un-noticed and I think he has a point (his second paragraph, after Macrakis' and my comment). NikoSilver 16:18, 24 February 2006 (UTC)

Loanword versus English words of Greek origin

I just realized that we have two similar articles Loanword and English words of Greek origin.

Loanword as the name implies, refer only to the immediate donor, hence the 5.32% figure from Scheler.

Now English words of Greek origin, also as the name implies, should include all of the above categories (indirect and coinages).

This is the basic difference between these two articles, therefore they should not include the same figures. Otherwise there wouldn't be any reason to have two articles. 21:45, February 24, 2006 Odysses

The articles do not include the same figures now. We also mention the 5.32% thing for clearing things out. What is your point? NikoSilver 23:10, 24 February 2006 (UTC)

5% is a joke

I spoke to a friend who is an etymologist and he has been living in Athens for 30 years. He said that %25 of English is made of Greek words. So it is nonsense to say %5. I got the best man to ask. :)

Somebody can even speak English by using greek words only and you can understand him very well. Very interesting isnt it? (unsigned comment by User:Tirigan 2006-04-08 15:50:43)

Did you read the article and the Talk discussion? The 5% and 25% figures have already been discussed. As for speaking English by using Greek words only, are you referring to that famous Zolotas speech?[33] It is no doubt an amusing and fun parlor trick but proves little, and anyway, many of the words he uses are either completely incomprehensible to an English-speaker ('synagonize'; 'eucharistia'; 'philoxenous': does appear in OED, but is very rare; 'trapeza': the root is only used for (circus) 'trapeze' and 'trapezoid') or are only used to refer to ancient Greek or Christian concepts ('archon'; 'kyrie'), or do not mean what he would like them to mean ('eulogize', used only for dead people; 'eugenic', used only for 'race improvement'); and most of the rest are at best unidiomatic. You could do a much better job if you restricted yourself to words of Germanic (in fact, people sometimes try to do that cf. Anglish) or Latin origin (this was popular in the 18th/19th centuries but is considered poor style today). On the other hand, it is true that you could not restrict yourself to words of Celtic or Slavic or Japanese origin. --Macrakis 18:12, 8 April 2006 (UTC)

I remember I had added a link to the speeches at the Xenophon Zolotas article. I think the above information is notable. Can you please add it there?  NikoSilver  ^{(T) @ (C)} 18:52, 8 April 2006 (UTC)

Well, I'll think about it. But I'm not sure how to do that without crossing the line into Original Research. --Macrakis 21:06, 8 April 2006 (UTC)

Simple: Use the definitions of an online etymological dictionary. I'd do it word by word, by replacing each word with the relevant link. For example:

I eulogize the archons of the Panethnic Numismatic Thesaurus...

By the way, "I don't see dead people" in eulogy!! Well, I think there are more important things in life, anyway...
 NikoSilver  ^{(T) @ (C)} 15:09, 11 April 2006 (UTC)

An etymological dictionary tells you about the history of the words, not the current meaning. A regular dictionary would be better, but doesn't capture idiom very well. Just ask some native speakers (who don't know Greek) what they think the speech means and whether it sounds like good English.... (but that is of course Original Research) --Macrakis 17:59, 11 April 2006 (UTC)

Hmmm... could you ever even smile, at least, Macrakis? But if you want to make it indeed serious: Is your point that people should continue to disregard what you appear to mostly admire (i.e. education) and just keep a short vocabulary? I too can see the difference between "eulogize" and "bless", I too can see that the whole text is exagerrated, but still, I am amazed to see what the late Zolotas managed to pull off! I am amazed to see that, indeed, under the right circumstances and with the proper education, a person can speak English, using only Greek words. The technicality of part of those words being abused (or unused) by uneducated or partly-educated people (like me-compared to you), is obsolete. My point in this is that this should be considered "good English" and not the other way round. A logical approach to this conversation would require us to exchange positions!  NikoSilver  ^{(T) @ (C)} 21:35, 11 April 2006 (UTC)

Niko, Zolotas's talk is amusing -- mostly for us bilinguals -- but it is just word-play, much as is A Void, which avoids that common sign which follows a, b, c, d. This 'shows' that that sign is not critical, right? so now stop using it! :-) --Macrakis 02:04, 12 April 2006 (UTC)

Check the sentences: Endless spree the letter E. Free men (Greeks etc here) see the need. Hellenes feel the bender. Been there, thee? Cheers!!! :-) Me! 12/04/2006

Introductory paragraph

Why is there so much text before the first heading? I'm tempted to put a cleanup flag on the article. Garrie 05:12, 17 August 2006 (UTC)

from which greek dialect?

"or from Ancient Greek texts" -> from which dialect? I think it would be nice to explicit this, although I can't do it because I don't know the answer (and I can't find it on the web either). I think it's relevant because some characters were transcribed differently in different dialects, for instance glossa / glotta. I know English has both of these, but is there a dialect that is predominant? If not, it could be equally interesting to mention something like "from every Ancient Greek dialect". PeterMellow 06:22, 9 November 2006 (UTC)

more on Aristidis Konstantinidis

Since I'm writing an essay for a Pan-Hellenic Scholarship foundation, I did much research on this and found an article in Greek here: [34] For those who cannot read Greek, the Google translation of this is below, although, having had it personally translated over the phone by someone who is completely fluent in both languages, I must say this is a poor translation. Hopefully you will get the jist, though. Basically, Mr. Konstantinides says that 68% of all English words come from Greek, excluding botany, chemistry and zoology. If we add these, AND include words that are half of Greek origin, and half of another, such as Latin or something else, the total is 74.48%. I know this is really high, and it is definitely specialized, as many more scientific words come from Greek than do everyday words, but I believe it reflects the TOTAL English language, not just spoken English. For my report, this is a good number, although I understand that a Wikipedia article probably should be lower and reflect the percentage in everyday spoken English. I'm not good at Wikipedia formatting, so I'll try to sign my name. Below is the article, run through google translator. Sorry for the bad formatting...I am horrible at Wiki-style websites and understanding formatting...someone please edit the format of the below article. TheotokosFan (talk) 23:42, 29 July 2010 (UTC)

(removed copyvio)

Please don't cut and paste long articles into Wikipedia. It is a violation of the original author's copyright, and serves no purpose -- instead, you can link to Google Translate, if you like, like this. --Macrakis (talk) 00:35, 30 July 2010 (UTC)

Coined words

Macrakis, we had exhaustive discussions on the issue on coined words in this page and we have come to a final agreement. They should be included in this article as per above references. I see no point in coming back again on this issue. Odysses (₪) 20:33, 25 June 2013 (UTC)

I would like more English words of Greek origin to be included "Music" "Comedy" Tragedy" "drama" "strategy" "politics" — Preceding unsigned comment added by 72.220.219.168 (talk) 07:20, 5 December 2014 (UTC)

There are more words of Greek origin didn't included!

More English words of Greek origin need to be included "Music" "Drama" "Comedy" "Tragedy" "Politics" "Strategy" — Preceding unsigned comment added by 72.220.219.168 (talk) 07:22, 5 December 2014 (UTC)

Hi

Hello he'll are you 49.184.19.133 (talk) 04:57, 14 February 2023 (UTC)