Marjorie Herbert’s paper, "A New Classifier-Based Morpheme in German Sign Language (DGS)," is the result of her undergraduate thesis work and fieldwork in Berlin, Germany. While studying abroad, she was influenced by a paper on pluralization in German Sign Language by Roland Pfau and Markus Steinbach (2005). This paper helped her realize that data she had been collecting revealed a different kind of grammaticalization: “Pfau and Steinbach have a typology paper of grammaticalization paths and they list a few modality specific ones that are exclusive to sign languages. If my analysis is correct, this is an example of a new grammaticalization path that hasn’t been written about before.”

But what does she mean by a new grammaticalization path? She explains, “Normally DGS uses reduplication, or repetition of sign, to mark the plural form, but Pfau and Steinbach say that the phonology of most signs in DGS don’t allow that marker to attach to them. There are different alternatives you can make use of, usually by way of a classifier construction, but that adds another semantic component to it; classifier constructions create a sentence instead of a simple word or phrase. What I noticed is that the signers I was working with had this semantic bleaching effect going on with some of their classifier use, what I’m calling this ‘new classifier-based morpheme’. It looks like a classifier but it has the semantics of a simple plural, and you only really see it with nouns whose phonologies block the simple plural, like BIKE. The form is similar to the classifier but the movement element and the semantics are different. It’s a classifier that has been taken out of the classifier system and is being grammaticalized as a new plural marker. As such, it represents a new, modality-specific grammaticalization path—from classifier to functional item.”

After completing her undergraduate thesis, Marjorie never expected this work might be publishable material. After coming to UM, Professor Acrisio Pires encouraged her to try to submit her work to a conference. She presented at the Penn Linguistics Conference 39 last March and over the summer they invited her to submit a proceedings paper. It was finally published this month in University of Pennsylvania Working Papers in Linguistics, Volume 22.1 (2016), after being in the review process since September. However, she’s not done yet! She is currently developing the paper into a journal-length article to submit to Sign Language and Linguistics.

Marjorie (and UM Linguistics) is extremely excited about her first publication!


In Marjorie’s own words:

“Like many other spoken and sign languages, German Sign Language (Deutsche Gebärdensprache, DGS) displays variation in the surface form of the plural marker. This allomorphy is phonological-ly conditioned, triggered by particular phonological properties of the lexical sign to which the plu-ral marker attaches. Since the overt realization of this marker is licensed to a small set of nouns in the language with the proper phonological properties, the majority of lexical signs are left bare, or zero-marked, when the simple plural is formed (Pfau and Steinbach 2005, 2006). According to Pfau and Steinbach (2005), there are a number of what they call “alternative pluralization strategies” available in DGS as a repair for this underspecification in the simple plu-ral, including classifier constructions, spatial localization, and number phrases. These construc-tions are available for use with any lexical noun, regardless of whether the canonical plural mark-er can be overtly realized with that noun. When used in conjunction with a zero-marked plural, they argue, these constructions serve as alternative means for expressing a plurality of referents. I propose that a new morpheme for plural marking is emerging in the language and that it is drawn from the classifier system already available in DGS. This new morpheme attaches only to nouns with phonological features blocking the realization of the canonical plural in some way. For this reason, I conclude signers are beginning to use the classifier-based morpheme as a repair for this underspecification.”