[Lingtyp] Cross-morpheme glosses

[Kate Lindsey]

I agree with Christian that the choice of interlinear glossing has a lot to
do with the audience, context, and what you wish to convey. I enjoy reading
interlinear texts that allow the reader to toggle between a simple or a
more complex gloss to provide the most accessibility for readers. This is
difficult in a static text, however.

As for how to gloss your examples, your pen example reminds me of how
authors often add a literal translation or paraphrase in square brackets
after the free translation to assist the reader. So I might suggest adding
[pen] in brackets after your truer to the morphology gloss
instr-repet~write (1). Alternatively, you could do this in the morphology
line and have the complex morphology in the square brackets (2). For your
thinking example, I might gloss such a prefix with an "unknown"
abbreviation like unk or der `unknown derivational affix' (1). I think (2)
is closer to what you've suggested, but I think (1) is much better.

[image: image.png]\

Kate
____________________
Kate L. Lindsey
Assistant Professor of Linguistics
Boston University
Boston, MA 02215
http://ling.bu.edu/people/lindsey


On Fri, Jul 11, 2025 at 11:04 AM Konstantin Henke via Lingtyp <
lingtyp at listserv.linguistlist.org> wrote:

> Dear all,
>
> glossing texts of agglutinative languages frequently leads to the
> situation where an orthographic word consisting of multiple transparent
> morphemes can be glossed using an English word, i.e. usually when
> derivative morphology is present, such as Takbanuaz Bunun *ispapatas*
> 'pen', which is derived from the root *patas* 'to write' as *is-pa~patas*
> [INSTR-REPET~write] 'thing to frequently write with'. When glossing longer
> texts, I'm often unsure whether to indicate the morphology or not, and I
> usually go by my own transparency judgement; i.e. if the gloss says 'INSTR-REPET~write'
> and the translation mentions a 'pen', then the reader might be able to put
> two and two together, but in other cases the derived meaning might have
> been conventionalized in a way that is no longer transparent from its
> (original) morphological composition, so writing *ispapatas* without
> indicating morpheme boundaries and just glossing it as 'pen' might be the
> better option.
>
> However, the boundary between derivational and inflectional morphology is
> of course not always clear (like locative or instrumental markers often
> having a nominalizing function in many Austronesian languages, as in the
> example above) and is thus not a good diagnostic of whether to provide
> morphological transparency in the glossing or not. Secondly, sometimes the
> linguist wishes to both indicate morphology *and* provide a gloss for the
> entire cross-morpheme meaning, i.e. both *is-pa~patas* (instead of
> *ispapatas*) and 'pen'.
>
> Additionally, sometimes the concrete function of some derivative affix
> might not be known, but the meaning of the entire derived word is known,
> while the linguist wishes at the same time to indicate the presence of said
> affix. E.g. Bunun *miliskin* 'to think' consists of *mi- *(unclear)
> and the root *liskin* 'believe, think, assume, consider' (De Busser,
> 2009: 573). In a text, I'd like to indicate both the fact that *mi-* is a
> prefix and the fact that *miliskin* is best translated as 'to think'.
>
> I have thought about using the dot that usually has the opposite function
> (joining multiple meanings/functions/glosses to correspond to one morpheme)
> in order to join multiple morphemes to correspond to one gloss, i.e.
> writing *mi.liskin* with the gloss 'think' or *is.pa.patas* 'pen'. This
> does not seem to be common practice and might not actually be very useful
> since the morphemes themselves aren't glossed, but at least it would point
> to the fact that affixal morphology present instead of just giving the
> reader a "chunk" like *ispapatas* for them to figure it out themselves.
>
> I was wondering if someone could point me to the common practice or even
> recommendations like the Leipzig glossing rules (which do not treat this
> case) or even to an entirely different approach I might not have thought of.
>
> Thank you!
>
> Konstantin Henke
>
> –––
> De Busser, R. (2009). *Towards a grammar of Takivatan Bunun: Selected
> topics* (Doctoral dissertation, L).
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/lingtyp
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250712/218b47da/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 57720 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250712/218b47da/attachment.png>