[Lingtyp] Question about helpful design of a comparative, multilectal grammar

Rasmus Bernander rasmusbernander at gmail.com
Tue Jun 19 16:05:13 UTC 2018

Dear all,

I’m overwhelmed by the amount and the quality of the responses so far! Many
interesting references also. I still have to check out some of them more
properly (and find some of them!), but I think Evan (2003) in his
introduction really argues well both for some “whys” and “hows” with regard
to multilectal grammars (or “pan-dialectal grammars”, although for our case
the varieties are not perceived as different dialects but different
languages). And of course, there are many challenges beyond data
presentation which a multilectal grammar shares with a monolectal one, e.g.
regarding reliability and validity. We absolutely want to connect our
description to an archive of some kind and/or publish it electronically
online, not least for the benefit for the speakers themselves. Many thanks
for pointing out some literature on this matter. Many thanks also for
reminding us to take the very important issue of colorblindness into
consideration and to make sure to always coherently account for the
realization of a specific form/function in ALL languages. To avoid that,
perhaps one solution (which I’m personally very fond of) could be
introductory tables listing the various realizations of a given paradigm.
For example, the one on copula constructions could like below (SM= Subject




















* table ? affirmative copula constructions (in different temporal contexts)*

Starting the section on copula constructions with such a table, one would
make sure to have represented all form-meaning pairs of all languages.
However, in the discussion one wouldn’t necessarily have to illustrate,
say, the past tense copula construction for all languages. It would suffice
with one example of the “general” construction in contrast with the
exceptional Ishenye variant. In addition, comparative tables like this one
facilitates historical inferences, i.e. what are shared retentions and
shared innovations contra individual developments, and, in that case, what
are the plausible direction(s) of change. For example, it would seem that
the past tense form with an unspecified vowel <V> in Ikoma and Ngoreme
originates from a weakened °-*a*- (which resonates nicely with the fact
that there’s a past morpheme *-*a*- reconstructed for Proto-Bantu). This, I
believe, also strengthen the conclusions about the individual varieties.
Would you agree?

Best, Rasmus

PS. I would also like to take this opportunity to draw your attention to
two workshops/conferences devoted to issues like the one we're discussing
here (by accident, one takes place at my old department and the other one
at my new department!).

Fieldwork: Methods and Theory
Short Title: FMT

Date: 13-Dec-2018 - 14-Dec-2018
Location: Gothenburg, Sweden
Contact Person: Eva-Marie Bloom Ström
Meeting Email: fmt1[image:
Web Site:

Descriptive Grammars and Typology: The Challenges of Writing Grammars of
Underdescribed and Endangered Languages

Date: 27-Mar-2019 - 29-Mar-2019
Location: Helsinki, Finland
Contact Person: Arja Hamari
Meeting Email: grammarwriting2019 at helsinki.fi
Web Site:

2018-06-18 19:03 GMT+02:00 Claire Bowern <clairebowern at gmail.com>:

> I've used pan-dialectal grammars but haven't written one. A problem
> with all the grammars I've read is that, when there is no explicit
> mention of dialectal difference -- it's impossible to know how much to
> generalize. That is, if the grammar is about dialects X, Y, and Z, and
> a section mentions only X and Y, does it also apply to Z and the
> author didn't mention it? Or does Z have a different construction,
> covered elsewhere?
> Claire
> On Mon, Jun 18, 2018 at 3:03 AM Sebastian Nordhoff
> <sebastian.nordhoff at glottotopia.de> wrote:
> >
> > Dear Rasmus,
> > this is a very interesting topic. The most important principle to follow
> > would be, in my view, the separation of content and presentation.
> >
> > This means that your text should code which passages refer to Ikoma,
> > Nata, Isenye, Ngoreme, respectively, but it should NOT say how this is
> > to be represented visually.
> >
> > This will allow you to adapt the presentation of your content for
> > different audiences later on.  Depending on the audience, you could opt
> > for color coding, borders, fonts for the different varieties, or no
> > visual distinction at all.
> >
> > This kind of semantic coding is the standard way of doing things in
> > HTML+CSS or LaTeX, but you can also use MS Word styles to achieve this.
> >
> > One fundamental question would be whether the outcomes of this project
> > are to be represented in book form (linear representation), or whether a
> > non-linear approach (think Wikipedia) might be more useful. There are
> > arguments for both approaches.
> >
> > In addition to the 2012 book which Peter Austin mentioned, you might
> > find the following article interesting:
> > Nordhoff, Sebastian. 2008. Electronic reference grammars for typology:
> > Challenges and solutions. Language Documentation & Conservation
> > 2(2):296–324.
> > https://scholarspace.manoa.hawaii.edu/handle/10125/4352?mode=full
> >
> > Best wishes
> > Sebastian
> >
> >
> >
> >
> >
> >
> >
> > On 06/15/2018 03:02 PM, Rasmus Bernander wrote:
> > > Dear members of the Lingtyp list,
> > >
> > > I’m currently involved in a project called “Linguistic Variation as an
> > > Indicator of Historical Relations and Language Contact: A Comparative
> > > grammar of four Mara Bantu languages (Tanzania)”. The project is
> funded by
> > > Koneen Säätiö and led by Dr. Lotta Aunio, Department of Languages,
> > > University of Helsinki. As implied in the title, the project aims at
> > > offering a linguistic description of four closely related (yet
> structurally
> > > versatile) Bantu varieties, Ikoma, Nata, Isenye and Ngoreme (known
> > > collectively as the Western Serengeti languages).  More information
> about
> > > the project can be found at this homepage:
> > > https://blogs.helsinki.fi/mara-project/
> > >
> > > We are struggling a bit with the question about the ideal way of
> designing
> > > the linguistic description. We would like to ask you ”Humans who read
> > > grammars”, i.e. you researchers who make typological (and/or
> comparative
> > > and/or specific theoretical) work and thus have great experience in
> reading
> > > grammars as well as extracting information from grammars: What would
> you
> > > consider being the most helpful and straightforward way to organize the
> > > structure of a multilectal grammar of this kind? We would prefer to
> find a
> > > way to systemize the data in a manner where we don’t have to
> prioritize one
> > > variety over the others and where we can also present the subsystems
> of the
> > > non-main varieties in a coherent way. It seems that some grammars use
> > > color/symbol coding for different varieties. Do you consider that
> helpful?
> > > Or do you have other, similar ideas on how one would succeed in
> creating a
> > > really clear and comprehensible comparative grammar?
> > >
> > > Many thanks in advance!
> > >
> > > /Rasmus Bernander
> > >
> > >
> > >
> > > _______________________________________________
> > > Lingtyp mailing list
> > > Lingtyp at listserv.linguistlist.org
> > > http://listserv.linguistlist.org/mailman/listinfo/lingtyp
> > >
> > _______________________________________________
> > Lingtyp mailing list
> > Lingtyp at listserv.linguistlist.org
> > http://listserv.linguistlist.org/mailman/listinfo/lingtyp
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20180619/60053724/attachment.htm>

More information about the Lingtyp mailing list