Typical percentage difference between the source and target text in your language combination Thread poster: Kieran Dobson
|
|---|
I'm quite curious to know more about the differences between the word counts for source texts and target texts in different language combinations. I've noticed recently that the translations I do (Portuguese to English) generally result in something around an extra 5% (give or take a couple percent) in the target (English) text. From what I have read, I was led to believe that Romance languages are generally "wordier" than English, so my experience seems to be contrary to this. I had been thinki... See more I'm quite curious to know more about the differences between the word counts for source texts and target texts in different language combinations. I've noticed recently that the translations I do (Portuguese to English) generally result in something around an extra 5% (give or take a couple percent) in the target (English) text. From what I have read, I was led to believe that Romance languages are generally "wordier" than English, so my experience seems to be contrary to this. I had been thinking that the extra words were the result of me trying to improve clarity and remove ambiguities when preparing the target text, which could very well be the case, but, regardless, I think it'd be an interesting exercise to compare with other translators.
So, do other Portuguese to English translators see a similar percentage difference? What sort of differences do English to Portuguese translators see? What percentage differences are typically seen for other Romance language combinations (which I imagine may behave similarly to my language pair, in terms of word count similarities/differences)? And, what about for all the other language combinations (e.g., German, Arabic, Persian, Chinese, Tupi, whatever)? Especially considering this last question, I'm also interested to know which language combinations are typically charged according to: the target text, the source text, the number of characters (in source or target text), or another system. Some input from an agency point of view would be interesting to hear; for example, the reasoning why a "word count" is mostly used in translations between European languages (is it due to being more customer friendly/CAT friendly/business friendly/translator (un)friendly?), as opposed to a character count, which I've always thought would be more accurate and fairer in almost all instances. ▲ Collapse
| | | | | I'm surprised you say English is more wordy than Portuguese. | Jun 13, 2016 |
I generally find that in Spanish to English translation, the English is considerably shorter.
| | | | Roni_S 
Slovakia
Local time: 00:20
Slovak to English
I find that my translations from Slovak to English result in more words in English, but I chalk that up to the use of prepositions and articles, which Slovak makes little use of.
| | | | Sheila Wilson
Spain
Local time: 23:20
Member (2007)
English
+ ...
| English shorter than French | Jun 13, 2016 |
I always find my translations end up shorter than the source texts, probably by a quarter or maybe even a third. But some individual sentences might be longer, if they refer to French places, organisations etc. as they will have to include an explanation of sorts.
I don't think it's particularly important how we come to the rate as long as we're happy with what we get paid for the job. I like per word as a rate but maybe it's just because I'm used to it. As with changing currencies,... See more I always find my translations end up shorter than the source texts, probably by a quarter or maybe even a third. But some individual sentences might be longer, if they refer to French places, organisations etc. as they will have to include an explanation of sorts.
I don't think it's particularly important how we come to the rate as long as we're happy with what we get paid for the job. I like per word as a rate but maybe it's just because I'm used to it. As with changing currencies, you don't know the cost of anything for a while but you soon get used to it. ▲ Collapse
| | |
|
|
|
Lingua 5B
Bosnia and Herzegovina
Local time: 00:20
Member (2009)
English to Croatian
+ ...
| Depends on the type of text. | Jun 13, 2016 |
Sometimes English tends to use more compact and shorter "solutions" to express certain things, but not always. In the variants of my native language there can be dialects that prefer to complicate things out so to speak and express the same concept using a longer word sausage.
Example:
Croatian/Bosnian: tlakomjer
Serbian: aparat za merenje krvnog pritiska
English: blood pressure monitor
In regards to this specific case, there was a Bosnian do... See more Sometimes English tends to use more compact and shorter "solutions" to express certain things, but not always. In the variants of my native language there can be dialects that prefer to complicate things out so to speak and express the same concept using a longer word sausage.
Example:
Croatian/Bosnian: tlakomjer
Serbian: aparat za merenje krvnog pritiska
English: blood pressure monitor
In regards to this specific case, there was a Bosnian doctor dealing with a Serbian nurse. Nurse, can you please pass me tlakomjer? Nurse: What is tlakomjer? He explains. Oh she says, you mean "aparat za merenje krvnog pritiska?". Yes, doctor says, but by the time I pronounce that, my patient's condition can get even worse. : D
I would say that often times English can go short and compact in ads and marketing and/or it tends to functionally simplify words and compounds more than my language does. ▲ Collapse
| | | | | What exactly is your point? | Jun 13, 2016 |
Kieran Dobson wrote:
I'm quite curious to know more about the differences between the word counts for source texts and target texts in different language combinations. I've noticed recently that the translations I do (Portuguese to English) generally result in something around an extra 5% (give or take a couple percent) in the target (English) text. From what I have read, I was led to believe that Romance languages are generally "wordier" than English, so my experience seems to be contrary to this. I had been thinking that the extra words were the result of me trying to improve clarity and remove ambiguities when preparing the target text, which could very well be the case, but, regardless, I think it'd be an interesting exercise to compare with other translators.
So, do other Portuguese to English translators see a similar percentage difference? What sort of differences do English to Portuguese translators see? What percentage differences are typically seen for other Romance language combinations (which I imagine may behave similarly to my language pair, in terms of word count similarities/differences)? And, what about for all the other language combinations (e.g., German, Arabic, Persian, Chinese, Tupi, whatever)? Especially considering this last question, I'm also interested to know which language combinations are typically charged according to: the target text, the source text, the number of characters (in source or target text), or another system. Some input from an agency point of view would be interesting to hear; for example, the reasoning why a "word count" is mostly used in translations between European languages (is it due to being more customer friendly/CAT friendly/business friendly/translator (un)friendly?), as opposed to a character count, which I've always thought would be more accurate and fairer in almost all instances.
What are you suggesting? To charge by which ever text is longer? It doesn't work like that. No two languages can be translated 1 on 1. I always charge by source text, unless it I can't counted, because of the format. So, make up your mind. The way I see it, you win some, you loose some, but be clear to your client, and you profit the most.
| | | | | Word count is only little more, but words are longer. | Jun 13, 2016 |
In my pair English to German, experience shows that the word count is only slightly higher in German than in English, but the German average word length is higher, therefore the page count is considerably higher in German. Also, character restrictions are a problem - for example when a translation is about localising an app and the client specifically asks for impossibly short words. For the second time now, someone asked me to translate the word "buy" to German but PLEASE not to exceed the orig... See more In my pair English to German, experience shows that the word count is only slightly higher in German than in English, but the German average word length is higher, therefore the page count is considerably higher in German. Also, character restrictions are a problem - for example when a translation is about localising an app and the client specifically asks for impossibly short words. For the second time now, someone asked me to translate the word "buy" to German but PLEASE not to exceed the original word length, because it has to fit on a button. Go figure. ▲ Collapse
| | | | Lingua 5B
Bosnia and Herzegovina
Local time: 00:20
Member (2009)
English to Croatian
+ ...
| A better question yet would be... | Jun 13, 2016 |
do 10k source words in non-native English and 10k source words in native English have the same value?
Honestly with non-native source texts you often have to stop and think now "what is this", so it may slow down the translation process significantly. My point is, it's not always about the word-count, there are other factors at play.
| | |
|
|
|
Georgia Morg (X)
United Kingdom
Local time: 23:20
Portuguese to English
| Portuguese to English | Jun 13, 2016 |
I find, like you do, that the word count in English is at least 5% longer. I think it´s all the prepositions and auxiliary verbs!
| | | | neilmac
Spain
Local time: 00:20
Spanish to English
+ ...
philgoddard wrote:
I generally find that in Spanish to English translation, the English is considerably shorter.
Same here. Off the top of my head, I'd say that most Spanish texts I handle seem to be about 10-20% wordier in general, although there are always exceptions to the rule.
PS: I do tend to exaggerate in general, so perhaps 5-10% may be a more accurate figure, but hey...
[Edited at 2016-06-14 09:21 GMT]
| | | | | French and Spanish to English | Jun 14, 2016 |
I find, as a rule, that my English version of a French or Spanish text is slightly shorter.
There are exceptions. For example, "statuts" and "estatutos" mean "articles of association" (three words instead of only one) so, if that expression occurs frequently in the text, the English version may be a bit longer. I prefer to charge according to the source word count (and so do most clients) but if the source text can't be reliably counted, I charge according to the target count (with the cli... See more I find, as a rule, that my English version of a French or Spanish text is slightly shorter.
There are exceptions. For example, "statuts" and "estatutos" mean "articles of association" (three words instead of only one) so, if that expression occurs frequently in the text, the English version may be a bit longer. I prefer to charge according to the source word count (and so do most clients) but if the source text can't be reliably counted, I charge according to the target count (with the client's prior agreement). ▲ Collapse
| | | | | Depends on the writer | Jun 14, 2016 |
Scandi translators like to say English translations are 20-25% longer because of compound words, but my translations are normally only 0-5% longer, presumably because I do like to cut the crap.
EitherthatorIwritewithoutspacesquitealot.
But ultimately it will always depend on how economically the author and the translator write.
As for pricing - it makes no difference whether you charge by source or target. Your word rate should reflect the amount of work, ... See more Scandi translators like to say English translations are 20-25% longer because of compound words, but my translations are normally only 0-5% longer, presumably because I do like to cut the crap.
EitherthatorIwritewithoutspacesquitealot.
But ultimately it will always depend on how economically the author and the translator write.
As for pricing - it makes no difference whether you charge by source or target. Your word rate should reflect the amount of work, which is the same. ▲ Collapse
| | |
|
|
|
Daryo
United Kingdom
Local time: 23:20
Serbian to English
+ ...
| counting characters | Jun 14, 2016 |
is far better metric - you have no possible arguments about what is a "word", like compound words being or not one or two or three words, for example)!
Another aspect of this question, maybe not intended when it was asked: the proportion of the character counts between the translation and the source text can be indirectly a rather good indicator of quality!
Namely if the average increase in character count from language A to language B is say 20% (yes, there are languag... See more is far better metric - you have no possible arguments about what is a "word", like compound words being or not one or two or three words, for example)!
Another aspect of this question, maybe not intended when it was asked: the proportion of the character counts between the translation and the source text can be indirectly a rather good indicator of quality!
Namely if the average increase in character count from language A to language B is say 20% (yes, there are language combinations where it's normal), any translation of more that few pages than manages to inflate the character count by say 40% is very likely to full of pointless verbosity.
As for charging, what should be reflected is the amount of work involved - which can lead to rather complicated estimations, and one of the aspects is that counting characters in the translated text could make sense if the ST is full of abbreviations that need first to be expanded.
. ▲ Collapse
| | | | EvaVer (X)
Local time: 00:20
Czech to French
+ ...
| I mostly count in characters, not words | Jun 14, 2016 |
and my findings are that Latin languages (in my case, French and Romanian) are indeed longer than English, Czech and Slovak (which are about the same), on average by 20%. Bulgarian is surprisingly long in terms of characters, but probably wouldn't be in terms of words (long words, post-positive articles). As others said, it depends on text type - the diference (FR/CZ, SK or EN) is up to 40% for letters (wordy and over-polite French style). English legalese is also longer than CZ or SK, which are... See more and my findings are that Latin languages (in my case, French and Romanian) are indeed longer than English, Czech and Slovak (which are about the same), on average by 20%. Bulgarian is surprisingly long in terms of characters, but probably wouldn't be in terms of words (long words, post-positive articles). As others said, it depends on text type - the diference (FR/CZ, SK or EN) is up to 40% for letters (wordy and over-polite French style). English legalese is also longer than CZ or SK, which are more matter of fact. BUT, translators that are not so good tend to have target text longer than source text whatever the language pair - I have always said, if your target sentence is much longer than the source, you have it wrong - dump it and start from scratch. Of course, the cause can reside in acronyms than you must explain in the target language. ▲ Collapse
| | | | Kieran Dobson
Brazil
Local time: 19:20
Portuguese to English
TOPIC STARTER | Quantifying the quantifiables more accurately | Jun 14, 2016 |
Robert Rietvelt wrote:
What are you suggesting? To charge by which ever text is longer? It doesn't work like that. No two languages can be translated 1 on 1. I always charge by source text, unless it I can't counted, because of the format. So, make up your mind. The way I see it, you win some, you loose some, but be clear to your client, and you profit the most.
As some of the other respondents have indicated, the ideal situation would be to always charge for the amount of time involved in your work. But even this involves a lot of different variables such as level of difficulty of the source text's subject matter (subjective), format of the text (which formats are more difficult and time consuming to translate, and how much more difficult?), and the most easily quantifiable: word count/character count. The discrepancies in a word count/character count are probably much more evident when you compare different subjects/areas. In order to quickly illustrate this, I selected two consecutive passages of 200 words from the English version of a literary translation, which gave a character count of 886 and 959 characters (8% MORE), respectively. I then chose two consecutive 200 word passages from the English version of a chemistry text, which gave a character count of 1396 and 1267 characters (9% LESS), respectively. The difference between the average character count of the 200 word literary text and the 200 word chemistry text was a massive 44%! Small sample size, yes; however, I guess the answer to your question is: If something CAN be quantified, do it as accurately as possible (i.e., choose a character count over word count). For my direct clients, I quote according to the character count in the source text.
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Typical percentage difference between the source and target text in your language combination | Protemos translation business management system | Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!
The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.
More info » |
| | Wordfast Pro | Translation Memory Software for Any Platform
Exclusive discount for ProZ.com users!
Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value
Buy now! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news
