FUNKNET Digest, Vol 99, Issue 3

Spruiell, William C sprui1wc at cmich.edu
Thu Dec 8 19:56:27 UTC 2011


This would be highly artificial but... could you simply use an
*orthographic* marker -- one that isn't intended to stand for a linguistic
unit at all? That would partially avoid the "pattern-distortion" issue,
and if it's something uncomplicated, like a <^>, it wouldn't be
particularly distracting for readers who didn't need it in context.

Bill Spruiell
Central Michigan University



>
>Message: 1
>Date: Thu,  8 Dec 2011 11:59:00 +0200
>From: john at research.haifa.ac.il
>Subject: [FUNKNET] Standardizing relativization in Dinka (and other
>	languages?)
>To: Richard Hudson <dick at ling.ucl.ac.uk>
>Cc: funknet <funknet at mailman.rice.edu>
>Message-ID: <1323338340.4ee08a6488098 at webmail.haifa.ac.il>
>Content-Type: text/plain; charset=windows-1255
>
>Dear Funknetters,
>I'm currently working with some Dinka speakers one how their orthography
>and
>writing system in general might be improved (the present system is clearly
>inadequate--even native speakers who wrote one of the Bible translations
>can't read their own translation at all fluently). There are many
>problems, but
>one of them seems to be that there is no standardized efficient way to
>make
>relative clauses--there seem to a wide variety of ad-hoc tactics which can
>be understood correctly but only with a lot of work. Some relative
>clauses are
>formally identical to sentences while others use morphemes which have a
>wide
>variety of other functions (articles, demonstratives, the 'be' verb,
>personal 
>pronouns, and prepositions). I know that 'that' can introduce relative
>clauses
>and also be a demonstrative adjective and a demonstrative pronoun but
>this is
>much much worse. The problem is even more serious because they use
>relatively
>few nominalizations but instead use something which  looks like a relative
>clause (e.g. 'Jesus' disciples' is translated every time as if it were
>'the men
>who were following Jesus'). I'm even finding that when I'm reading myself
>I
>mostly identify relative clauses by the head noun which often literally
>means
>'person' or 'thing'--for example, 'raan' and 'mony' in principle are both
>translated as 'person', but 'raan' is very often associated with something
>which would translate as a relative clause while 'mony' isn't. I don't
>really
>know what to do with this. I'm thinking of suggesting to them that some
>standardized ways to make relatives have to be chosen and stuck to. Do
>any of
>you have experience with anything like this?
>Thanks,
>John
>
>
>
>------------------------------------------------------------------------
>This message was sent using IMP, the Webmail Program of Haifa University
>
>
>------------------------------
>
>Message: 2
>Date: Thu, 08 Dec 2011 09:22:33 -0600
>From: "Mike Cahill" <mike_cahill at sil.org>
>Subject: Re: [FUNKNET] Standardizing relativization in Dinka (and
>	other	languages?)
>To: "Richard Hudson" <dick at ling.ucl.ac.uk>
>Cc: funknet <funknet at mailman.rice.edu>
>Message-ID: <80543a5f6d94884c8665f69ef4615cb9 at sil.org>
>Content-Type: text/plain;	charset="us-ascii"
>
>Hi John,
>
>Two issues here: how Dinkas do relative clauses, and the apparent
>difficulty in reading them.
>
>The first line of investigation I'd take is looking at natural texts in
>Dinka, and how relative clauses are naturally used. It may be that the
>translators didn't use the most appropriate form of relative clause for a
>particular context, and that in itself would reduce reading fluency - the
>reader is hit with an unexpected way of expressing things. I'd also note
>that different genres of texts may have quite different types of expected
>relative clauses. Narratives and hortatory texts should be examined
>separately, as a minimum. Unfortunately, to get a reasonable answer on
>this is going to take a fair amount of work.
>
>For your specific example, "disciple" is one of those terms that often
>doesn't have a one-word equivalent in a local language. Does it in Dinka?
>It sounds like it may not. If not, then the translator needs to unpack
>the meaning, and unfortunately for your frustration level, it may be that
>the most natural way to express the concept is with a relative clause.
>Maybe; I don't know Dinka!
>
>I'd be VERY leery of advocating a single "standardized efficient" way of
>presenting relative clauses, or any other syntactic or discourse
>structure. It's likely to distort the language's natural patterns.
>Variation is usually there for a reason, even if native speakers can't
>articulate what that might be (as most English speakers can't either).
>
>Mike Cahill
>
>
>-----Original Message-----
>From: funknet-bounces at mailman.rice.edu
>[mailto:funknet-bounces at mailman.rice.edu] On Behalf Of
>john at research.haifa.ac.il
>Sent: Thursday, December 08, 2011 3:59 AM
>To: Richard Hudson
>Cc: funknet
>Subject: [FUNKNET] Standardizing relativization in Dinka (and other
>languages?)
>
>Dear Funknetters,
>I'm currently working with some Dinka speakers one how their orthography
>and writing system in general might be improved (the present system is
>clearly inadequate--even native speakers who wrote one of the Bible
>translations can't read their own translation at all fluently). There are
>many problems, but one of them seems to be that there is no standardized
>efficient way to make relative clauses--there seem to a wide variety of
>ad-hoc tactics which can be understood correctly but only with a lot of
>work. Some relative clauses are formally identical to sentences while
>others use morphemes which have a wide variety of other functions
>(articles, demonstratives, the 'be' verb, personal pronouns, and
>prepositions). I know that 'that' can introduce relative clauses and also
>be a demonstrative adjective and a demonstrative pronoun but this is much
>much worse. The problem is even more serious because they use relatively
>few nominalizations but instead use something which  looks like a
>  relative clause (e.g. 'Jesus' disciples' is translated every time as if
>it were 'the men who were following Jesus'). I'm even finding that when
>I'm reading myself I mostly identify relative clauses by the head noun
>which often literally means 'person' or 'thing'--for example, 'raan' and
>'mony' in principle are both translated as 'person', but 'raan' is very
>often associated with something which would translate as a relative
>clause while 'mony' isn't. I don't really know what to do with this. I'm
>thinking of suggesting to them that some standardized ways to make
>relatives have to be chosen and stuck to. Do any of you have experience
>with anything like this?
>Thanks,
>John
>
>
>
>------------------------------------------------------------------------
>This message was sent using IMP, the Webmail Program of Haifa University
>
>
>
>
>
>------------------------------
>
>Message: 3
>Date: Thu,  8 Dec 2011 18:43:19 +0200
>From: john at research.haifa.ac.il
>Subject: Re: [FUNKNET] Standardizing relativization in Dinka (and
>	other	languages?)
>To: Mike Cahill <mike_cahill at sil.org>
>Cc: Richard Hudson <dick at ling.ucl.ac.uk>, funknet
>	<funknet at mailman.rice.edu>
>Message-ID: <1323362599.4ee0e9273ecb8 at webmail.haifa.ac.il>
>Content-Type: text/plain; charset=windows-1255
>
>Hi Mike,
>My information is based completely on the translation of the Book of
>Matthew.
>Obviously I'll look at other texts, but there really aren't that many. I
>don't
>have direct evidence that relative clauses in particular are difficult to
>read--
>I don't know any Dinkas who are linguistically sophisticated enough to
>make
>such a statement (or could even identify what a relative clause is, for
>that
>matter), they just report general difficulty. I'm just making a guess
>based
>upon what I know about the language. I certainly wouldn't suggest a single
>formula, I just meant to have a certain set of formulas which they can
>use,
>ones which would limit processing difficulty. The problem isn't too much
>variation, the problem is too much confusion. I would assume that the
>difference between relative clauses and simple sentences is clear in
>spoken
>language through intonation, but this can't be directly represented in
>writing.
>I'm also not saying that there's an inherent problem with using
>relativization
>instead of nominalization, I'm just saying that this adds greatly to the
>number
>of relative-like constructions in texts.
>John
>
>
>
>
>
>
>Quoting Mike Cahill <mike_cahill at sil.org>:
>
>> Hi John,
>>
>> Two issues here: how Dinkas do relative clauses, and the apparent
>>difficulty
>> in reading them.
>>
>> The first line of investigation I'd take is looking at natural texts in
>> Dinka, and how relative clauses are naturally used. It may be that the
>> translators didn't use the most appropriate form of relative clause for
>>a
>> particular context, and that in itself would reduce reading fluency -
>>the
>> reader is hit with an unexpected way of expressing things. I'd also
>>note that
>> different genres of texts may have quite different types of expected
>>relative
>> clauses. Narratives and hortatory texts should be examined separately,
>>as a
>> minimum. Unfortunately, to get a reasonable answer on this is going to
>>take a
>> fair amount of work.
>>
>> For your specific example, "disciple" is one of those terms that often
>> doesn't have a one-word equivalent in a local language. Does it in
>>Dinka? It
>> sounds like it may not. If not, then the translator needs to unpack the
>> meaning, and unfortunately for your frustration level, it may be that
>>the
>> most natural way to express the concept is with a relative clause.
>>Maybe; I
>> don't know Dinka!
>>
>> I'd be VERY leery of advocating a single "standardized efficient" way of
>> presenting relative clauses, or any other syntactic or discourse
>>structure.
>> It's likely to distort the language's natural patterns. Variation is
>>usually
>> there for a reason, even if native speakers can't articulate what that
>>might
>> be (as most English speakers can't either).
>>
>> Mike Cahill
>>
>>
>> -----Original Message-----
>> From: funknet-bounces at mailman.rice.edu
>> [mailto:funknet-bounces at mailman.rice.edu] On Behalf Of
>> john at research.haifa.ac.il
>> Sent: Thursday, December 08, 2011 3:59 AM
>> To: Richard Hudson
>> Cc: funknet
>> Subject: [FUNKNET] Standardizing relativization in Dinka (and other
>> languages?)
>>
>> Dear Funknetters,
>> I'm currently working with some Dinka speakers one how their
>>orthography and
>> writing system in general might be improved (the present system is
>>clearly
>> inadequate--even native speakers who wrote one of the Bible translations
>> can't read their own translation at all fluently). There are many
>>problems,
>> but one of them seems to be that there is no standardized efficient way
>>to
>> make relative clauses--there seem to a wide variety of ad-hoc tactics
>>which
>> can be understood correctly but only with a lot of work. Some relative
>> clauses are formally identical to sentences while others use morphemes
>>which
>> have a wide variety of other functions (articles, demonstratives, the
>>'be'
>> verb, personal pronouns, and prepositions). I know that 'that' can
>>introduce
>> relative clauses and also be a demonstrative adjective and a
>>demonstrative
>> pronoun but this is much much worse. The problem is even more serious
>>because
>> they use relatively few nominalizations but instead use something which
>> looks like a relative clause (e.g. 'Jesus' disciples' is translated
>>every
>> time as if it were 'the men who were following Jesus'). I'm even
>>finding that
>> when I'm reading myself I mostly identify relative clauses by the head
>>noun
>> which often literally means 'person' or 'thing'--for example, 'raan' and
>> 'mony' in principle are both translated as 'person', but 'raan' is very
>>often
>> associated with something which would translate as a relative clause
>>while
>> 'mony' isn't. I don't really know what to do with this. I'm thinking of
>> suggesting to them that some standardized ways to make relatives have
>>to be
>> chosen and stuck to. Do any of you have experience with anything like
>>this?
>> Thanks,
>> John
>>
>>
>>
>> ------------------------------------------------------------------------
>> This message was sent using IMP, the Webmail Program of Haifa University
>>
>>
>>
>>
>
>
>
>
>------------------------------------------------------------------------
>This message was sent using IMP, the Webmail Program of Haifa University
>
>
>End of FUNKNET Digest, Vol 99, Issue 3
>**************************************



More information about the Funknet mailing list