Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

MWE spellings with 'v' instead of 'b' #421

Open
funderburkjim opened this issue Jan 16, 2019 · 29 comments
Open

MWE spellings with 'v' instead of 'b' #421

funderburkjim opened this issue Jan 16, 2019 · 29 comments

Comments

@funderburkjim
Copy link
Contributor

funderburkjim commented Jan 16, 2019 •
edited
Loading

A user (vaibhav.niku) suggested that in MWE, वाढं should be changed to बाढं.
In this case, I found that MW does have वाढ as an alternate spelling to बाढ; furthermore in MWE the 'vAQa' (slp1) spelling occurs 12 times, but the 'bAQa' spelling does not occur at all. Thus, I decided to leave the 'vAQa' spellings unchanged in MWE, even though the 'bAQa' spelling appears to be the more usual one.

Question: Agree with this choice?
@funderburkjim
Copy link
Contributor Author

Vaibhav's comment made be think of checking other words. I found these cases where 'vahu' needs to be changed to 'bahu' (much), and 'vAhu' needs to be changed to 'bAhu' (arm).

; vahuH -> bahuH
; vAhulyaM -> bAhulyaM
; vahuranDravAn -> bahuranDravAn
; vahuvistIrRaH -> bahuvistIrRaH
; vahulaH -> bahulaH
; vahusamunnayavAn -> bahusamunnayavAn
; vahuBiH -> bahuBiH
; udvAhuH -> udbAhuH
; vahuSirAH -> bahuSirAH
; vahupaMktiH -> bahupaMktiH
; vahuH -> bahuH
; vahuSo -> bahuSo
; vAhuyudDasambanDI -> bAhuyudDasambanDI
; vahumAnaM -> bahumAnaM
; vahuDanaH -> vahuDanaH
; vahuDA -> bahuDA
; vahurasaH -> bahurasaH
; suvahuH -> subahuH

These changes are needed since

  • 'vahu' and 'vAhu' are not alternates of 'bahu','bAhu', at least according to MW
  • There are many (700+) instances of the 'bahu','bAhu' spellings in MWE.

@funderburkjim
Copy link
Contributor Author

Other changes?

I couldn't think of other things to check for in this 'v/b' study of MWE.

Any suggestions?

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 16, 2019 via email

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 16, 2019 via email

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 16, 2019 via email

@ghost
Copy link

ghost commented Jan 17, 2019

Hi

I was hoping that some obvious explanation for the discrepancy would be found (typesetting trouble or scanning error). I did check MWE's Preface (as I had promised!) and found nothing about it. And this seems to be no scanning error, for another scan (at https://archive.org/details/in.ernet.dli.2015.553776 ) shows 'v's too.

The MWE dictionary was published in 1851, and the MW dictionaries were published much later (1872 and 1899). It would then seems that Monier-Williams thought that वाढं was the canonical spelling upto 1851, and changed his opinion sometime later. (Same for a lot many v words.)

I have no opinion on what should be done in this case! Youngsinn's point makes sense too. (Besides, for scholarly work, another thing to worry about: Where do you stop? A thought exercise: what if future scholarship shows that the 'v' words are to be preferred?)

Probably, some sort of footnote asking to check the corresponding 'b' words too would be the best.

If I come across more information, I will inform! If I can think of some way of identifying the suspicious words, I will also inform.

Many thanks for all the work to everyone involved! Works like these are the best thing about the internet!

@ghost
Copy link

ghost commented Jan 17, 2019 •
edited by ghost
Loading

Probably, some sort of footnote asking to check the corresponding 'b' words too would be the best.

Better: When the 'v' word is missing in MW, the MWE's entry could put the corresponding 'b' word in brackets. So:
Much: वहु [बहु], ...

This should satisfy both the purists (like Youngsinn above), and those who are using the dictionary for work.

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 17, 2019 via email

@gasyoun
Copy link
Member

gasyoun commented Jan 17, 2019 •
edited
Loading

Question: Agree with this choice?

Yes.

we are NOT supposed to change someone's book or dictionary
arbitrarily even if there are original errors (not digitization errors).
Whether it is variant spelling or not, _which is canonical spelling of
a certain word_s, etc. should be addressed in academic articles.

Disagree. But we document everything. No change is ever done just so.

@ghost
Copy link

ghost commented Jan 19, 2019

It is likely an issue with what is to be taken as the canonical spelling.

For 'exterior', MWE says:

EXTERIOR , a. वाह्यः -ह्या -ह्यं, ETC. -- all with 'v'.

MW72 lists "being outside" as meanings for both vAhya and bAhya, and it further adds in bAhya:

बाह्य bāhya, as, ā, am (fr. bahis, q. v. In classical Sanskrit this word is more usually written vāhya, but in the Veda bahis and bāhya are the usual forms, the nom. pl. bāhye following the pronominal declension), being outside, ETC.

[Please specially note "the usual forms" in the Vedas, i.e., vahis and vāhya are also present!]

MW [1899] does not list exterior as a meaning for vAhya, but does ask the reader to refer to bAhya too. In bAhya, he says:
बाह्य mf(आ)n. (fr. बहिस्; in later language also written वाह्य q.v.; m. nom. pl. बाह्ये, ŚBr. ) being outside, ETC.

As with MW [1899], PWG [1855] lists "being outside" only for bAhya. And unlike MW, PWG's vAhya does not refer the reader to bAhya.

By the way, the v-b interchange is quite common in Hindi too. (Many people pronouce my name as Baibhab -- though I suppose the interchange is considered a bit uncultured now.)

@gasyoun
Copy link
Member

gasyoun commented Jan 19, 2019

interchange is considered a bit uncultured now

No, it's about regions. Same as it was in Vedic - it's a different region than Classical Sanskrit. MW is a remake of PWG.

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 19, 2019 via email

@drdhaval2785
Copy link
Contributor

I am of the view that b and v should be left as they are.

For the record, there have been plenty of print errors corrected in MW over the past 8 years. https://github.com/sanskrit-lexicon/CORRECTIONS/blob/master/dictionaries/MW/mw_printchange.txt.

Just bringing it to the notice, because it is not the case that we have not corrected print errors before.

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 19, 2019 via email

@funderburkjim
Copy link
Contributor Author

funderburkjim commented Jan 19, 2019 •
edited
Loading

One difficulty especially pertinent to MWE is the quality of the scanned image of the original edition.
Namely, it is not very good. Thus it is often difficult to determine exactly what the printed spelling is.
For instance, the difference between devanagari 'b' and 'v' is quite small. Take a look at this from 'much' in MWE.
image

When I look at it, and try to say which is a printed 'b' and which a printed 'v', my eyes get crossed and I cannot really say for sure.

So in such cases, we must use other methods to decide what the digitization should contain. I term these ill-defined 'other methods' common sense, but there will surely be differences of opinion there.
For me, these common sense arguments are sufficient to make a change to the digitization for the
list starting with vahuH -> bahuH.

Do others agree for this list?

@Sonnetag and @vniku . Thanks for discussion points.

In general, our aim in making corrections is consistent with the strong view expressed by Youngsinn.

Also, when we knowingly make a change to the printed digitization, I try to be vigilant in making a reference to that change in a 'printchange' file. These printchange files are kept in this repository. For instance, here is link to mw_printchange.txt. [Addendum: @drdhaval2785 already mentioned this above. No harm in repeating.]

Other interested people ( I think @zaaf2 ) have suggested that somehow the information in these print change files should be incorporated into the displays; this is similar to the suggestion by @vniku above.

@ghost
Copy link

ghost commented Jan 20, 2019 •
edited by ghost
Loading

Largely offtopic:

interchange is considered a bit uncultured now

No, it's about regions.

Probably true. I realized that the people who pronounce my name as Baibhab systematically use b's for all v's.

The "uncultured" part is an interesting debate, and I'll add a few comments because @Sonnetag objected to it. In India, the Bengalis, never switched / switched back from b to v, even in literature. If I remember correctly, a century back they had a debate about whether to make the switch to bring it in line with Sanskrit (classical Sanskrit), and they decided against it.

So, for at least the Bangla words, they just use b, and nobody objects to it! So, the previous President of India was named 'Pranab Mukherjee' -- प्रणब for Hindi/Sanskrit प्रणव. (They didn't 'correct' any word. A prominent reporter is named 'Barkha Dutta' -- बरखा for Sanskrit वर्षा.)

So, I suppose things become "uncultured" when the local (regional) high culture does not support it! (b is considered "cultured" in Bengal, but "uncultured" in the neighbouring Bihar.)

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 20, 2019 via email

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 20, 2019 via email

@ghost
Copy link

ghost commented Jan 20, 2019

Sonnetag's suggestion about a prominent link to the page listing the corrections, on the dictionary's homepage is fine. Another suggestion for corrections:

I like the way nietzschesource.org handles original corrections. The word/phrase is corrected, but the word's background is coloured. On clicking the word, you can read what the originally printed word/phrase was. If you print the page, you'd just get the corrected word (no original word, no colour).

E.g. check the page for Also Sprach Zarathustra, Erster Teil. The printed sentence reads, "Einsamer, du gehst den Weg des Liebenden: dich selber liebst du ...". 'Selber' is coloured, and on clicking it, you learn, "Erratum:selbst lies:selber [and there's a link to the page listing the basis for corrections]".

@Sonnetag: Interesting comments about pronunciation! Good luck for your planned translation of the Rigveda! I like the book too.

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 21, 2019 via email

@gasyoun
Copy link
Member

gasyoun commented Jan 21, 2019

When I look at it, and try to say which is a printed 'b' and which a printed 'v', my eyes get crossed and I cannot really say for sure.

Right.

For me, these common sense arguments are sufficient to make a change to the digitization for the list starting with vahuH -> bahuH.

It's not about changing the entry. But changing the ways we get there, maybe even adding ghost-words, that are purely for getting there.

previous President of India was named 'Pranab Mukherjee' -- प्रणब for Hindi/Sanskrit प्रणव.

Interesting case.

b is considered "cultured" in Bengal, but "uncultured" in the neighbouring Bihar

Fun to read.

Whether you provide file that includes list and explanation about the technical issues involved in scanning of the low quality scanned image in the header (?) of the MWE on-line dictionary should be ok, as long as those are listed or linked in a way that they can never be confused with the original dictionary entries by the author.

@Sonnetag the scans are as they are. No explanation other than here, no need to put more attention to it. It's just about that proofing is harder.

Can you provide a click button that says Must Read

Are you ready to write it?

It would be too much work to provide link for every relevant entries, I guess?

No, technically it's an easy one. The question remains who will write it.

I like the way nietzschesource.org handles original corrections.

manchem

Indeed, not bad at all. But we have had quite many headword issues in the past, right @funderburkjim ? So not all cases are easy. And anyway there is only Jim in the backend. Are you eager to learn coding or to go through lists of suspected words, @Sonnetag and @vniku ? One day all everyday users should become contributors, right?

@ghost
Copy link

ghost commented Jan 22, 2019

For the corrections, I volunteer to write the code (whatever is decided here). (I can code.)

For the b-v issue specifically, I'd like to draw everyone's attention to this:

MW72 says that both bAhya and vAhya are used in the literature. (And the same must be the case with all the words in Jim's list in the second post -- M-W would not have used the v-words in MWE unless they appeared somewhere in literature.)

Böhtlingk and Roth and Monier-Williams (and everyone else who worked on the dictionaries) must have cared at least as much for consistency and completion as we do. So, why does PWG does not even refer to bAhya in the vAhya entry? (And similarly, in Jim's list above, why does MW-1899 does not even refer the readers to the b-words under the v-words? -- these v words were "his own words"!)

I found the discrepancy because when I didn't find a v-word, I checked for the word with a b (for this interchange seems "natural" in Hindi). Here's what I'm wondering: do B. and R. and M-W assume that Sanskrit scholars not finding v-words would think of checking b-words too, without any prompting? And if true, wouldn't just a note on the home page about this be not sufficient? I mean, we don't want to be holier than the pope! (On MWE's homepage, something like, "Some words which are listed with a v in MWE will probably be found with a b in other dictionaries".)

Finding the suspicious words Is technically easy enough: just find all MWE words with v, and then query MW for all these words one-by-one.

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 22, 2019 via email

@ghost
Copy link

ghost commented Jan 22, 2019

About MWE, the meat of the Preface is page 4.

@Sonnetag, I don't understand what you are objecting about. I am not asking for an edit of MWE -- my last post was specifically about why leaving them unchanged may make ample sense. In general, that MWE and MW are not consistent is a real issue.

Hindi obviously has not much to say about how things are in Sanskrit. I brought up the issue to merely add an extra piece of information, namely, that the modern Indian languages too show the same inconsitency (between b and v). I am not a Sanskrit scholar.

(A bit lesser hostility, please!)

@Sonnetag
Copy link
Collaborator

Sonnetag commented Jan 22, 2019 via email

@gasyoun
Copy link
Member

gasyoun commented Jan 22, 2019 •
edited
Loading

For the corrections, I volunteer to write the code (whatever is decided here). (I can code.)

You can Python @vniku ?

assume that Sanskrit scholars not finding v-words would think of checking b-words too, without any prompting?

Yes, that's what I assume as well.

And if true, wouldn't just a note on the home page about this be not sufficient?

No, because it's just one out of 100 of hard cases and by no way the most frequent one. It's place is in a FAQ.

Finding the suspicious words Is technically easy enough: just find all MWE words with v, and then query MW for all these words one-by-one.

Been done many years ago already.

http://gasyoun.github.io/ see my compilation of Rigveda. @Sonnetag I hardly understand why you do not use https://www.sanskrit-lexicon.uni-koeln.de/scans/GRAScan/2014/web/webtc/indexcaller.php

The following few examples show that PWG is incomplete and not well planned and organized.

Untrue.

But in MW, the compounded words are not displayed under the key word.

Untrue. It's so only in Basic search. Try List at least.

  1. अनुचर् not found in PWG even though listed under चर्

Because @funderburkjim has not yet made the list. Jim, it's about time in 2019 to have dhatus with upasarga not only in MW, but PWG as well.

@ghost
Copy link

ghost commented Jan 22, 2019

assume that Sanskrit scholars not finding v-words would think of checking b-words too, without any prompting?

"Yes, that's what I assume as well."

And if true, wouldn't just a note on the home page about this be not sufficient?

"No, because it's just one out of 100 of hard cases and by no way the most frequent one. Its place is in a FAQ."

Great. This is an important piece of information pertaining to Jim's second question above (whether others approve of the substitutions in his list). Given the above, I suppose you too are more unhappy than happy with Jim's changes above.

(Sonnetag and others: I have no opinion on the issue. I am not competent to have an opinion on the issue.)

@gasyoun:

You can Python vniku?

Only a bit. I am sure that I can "get the work done", but -- though dependent on the complexity of the work and the time I have to do it -- the result won't be pretty to look at! Nevertheless, I'd be happy to do it. I would be happier if someone guides me in doing it.

http://gasyoun.github.io/

This is great. I have a feature request though! A header to the page, naming the translators would be nice. (When others quote the translations, they'd like to name the translator too, apart from providing a link.)

@Sonnetag: Understood. You needn't worry. "In general, our aim in making corrections is consistent with the strong view expressed by Youngsinn." (Jim above.) Some corrections do exist, nevertheless (the link provided by drdhaval above), and gasyoun is wondering if the corrections can be made more visible and obvious. As I said, I too would like to see more visible corrections. And obviously, you would like to see more visible corrections.

@funderburkjim
Copy link
Contributor Author

Regarding correction display of 'Selber' in Also Sprach Zarathustra

This is indeed a nice solution.

For the corrections, I volunteer to write the code (whatever is decided here)

@vniku -- If you are serious about this, we should open a new issue devoted just to such implementation. Or maybe it should be in a separate repository within the sanskrit-lexicon project?

There are lots of details that would be relevant. Here are some that come to mind at the moment.
Probably Python would be used to create from mw_printchange.txt (and similar files) a
sqlite database that could be consulted by dictionary displays.
The underlying digitizations (e.g. mw.txt) would also need additional markup (exact formats to be designed) to point to the mw_printchange.sqlite file.
The actual display of such information would involve modifying the display programs (written in PHP) and probably adding some Javascript (to do the nice html tooltip of the Selber example; maybe we can reverse engineer the nietsche site (or even wikipedia) to accomplish this ).

@ghost
Copy link

ghost commented Jan 24, 2019 •
edited by ghost
Loading

If you are serious about this, we should open a new issue devoted just to such implementation. Or maybe it should be in a separate repository within the sanskrit-lexicon project?

Yes, please go ahead and give the issue a place of its own. I'll add my comments there.

@funderburkjim funderburkjim mentioned this issue Jan 24, 2019
Open
@funderburkjim funderburkjim mentioned this issue Jan 25, 2019
Open
@drdhaval2785 drdhaval2785 mentioned this issue Dec 20, 2020
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@gasyoun @drdhaval2785 @funderburkjim @Sonnetag