On data collection and standard languages – a conversation with Dr. Stefan Dollinger

This month, Language Sciences is pleased to discuss with Dr. Stefan Dollinger about his recent research as a historical sociolinguist. Read on to learn about non-dominant varieties of languages, and how data-collection intersects with the work of a linguist.

What does your work/research focus on?

What I do now is a bit different from how I started out. I began as a historical slash sociolinguist, foregrounding the linguistic data we all work with in one way or another. Over the years, though, I became more and more interested in social constructions and identity expressions and I found that quantitative data, against the current trends, does not always give us the best perspective on a topic. This led me to the study of the construction of languages and language varieties – in which we all have a stake, often without realizing it. As I was working a lot on non-dominant varieties, such as Canadian English, I began to reorient myself back to my own roots in Austria. At the same time the Dictionary of Canadianisms on Historical Principles was published in second edition (www.dchp.ca/dchp2), which highlights usage characteristic of Canada or any of its regions, while a lot of research has moved into pan-national directions. Does the “national”, besides personal identity constructions, have a place in translanguaging? I think it does, but the research often seems to chuck the baby out with the bathwater. National dimensions can be important for small nations without being nationalistic, but as a positive kind of self-identification. 

Can you tell us about your recent research? 

Certainly. For the past eight years I’ve been in conversation, that is I've been trying to enter a conversation, with colleagues in German dialectology. It began with me noticing how the international approach — centred in pluricentricity (meaning more than one codified standard) — is epistemologically incompatible with the new, apparently data-driven way that German was being modelled on: that linguistic behaviour alone would give you a delimitation of varieties and languages. I saw that approach as anti-pluricentric and thought "a duck needs to be called a duck” and put my argument in a 2019 short monograph with Routledge, entitled The Pluricentricity Debate. 

A couple of years later I published a trade book, in Austrian German, on the question of the standard of German in Austria that went through multiple editions. The reception in the field, however, was cold. So cold that I began looking into the legacy of German dialectology and I emerged, from the archives, with new facts about German linguists that were rehired after WW II, their roles in the Third Reich, and the belittling of Standard Austrian German after the war. I concluded saying that German dialectology seems to be bound by a “One Standard German Axiom” (OSGA) today. OSGA seems to have been a field-defining assumption since the Grimms’ days. While contemporary linguists have, of course, nothing to do with the Third Reich, they seem bound by the axiom and the hegemonic traditions of the field. I said, for instance, that we need to look at the data collected by these pan-German philologists before we work with it and that we need to assess the data for biases: what did not fit their view of “deutsch” back in the day? What is missing in the data. Awareness is not great in that area yet.

That sounds like a tough research! What were the difficulties and challenges you've gone through in the process? 

Yes, that’s true, but I feel it’s a conversation that we need to have, especially in the age of big data, about the biases in linguistics that we carry on since the nationalistic days or the late 19th and early 20th centuries. 

One difficulty is to make sense of the terms that some schools use: “pluri-areality” — what is that? It took me two years of research to confirm that the concept seems indeed empty. It’s undefined. All it means is that it negates pluricentricity (in my 2019 The Pluricentricity Debate). That term has not been defined in more than 30 years, but is now much in vogue in German linguistics. 

Another difficulty with historical information is to locate it. As a historical linguist by training, I - luckily - knew where to look. In the archives another difficulty in researching how language or varieties were construed and modelled is often to see what is *not* there.  For instance, what do data collections look like that were collected during the Third Reich or in a period when nations were construed differently (e.g. one united Germany, with Austria)? What does it mean when the present data custodians proudly tell you that the word “Hitler” does not occur in the data? My immediate response, knowing about these pan-German and NS-researchers, was: who went through the collection post 1945 and purged it from superficially embarrassing data points? That would also mean, however, that the real biases are still there: who was not considered to be a good representative of the German language? Bilingual Austrian Slovene speakers, Romani and Sinti and, of course, Jewish speakers were likely excluded. There is a need to dehegemonize the conceptual approaches that date back to the 19^th century in our field, which requires first historical study of what has been kept, quite often, under wraps. Not just in German, in linguistics in general and here more and more approaches are being seen, which is encouraging.

Challenges? (Laughter). How do you discuss full-blown incompatibilities in linguistic approaches? The biggest issues is probably that not everybody knows what academic freedom is or how to respect it. If you think about it, we don’t teach it. When, for instance, you inform a colleague of a new paper coming out, you send the article proofs in collegial spirit, and they subsequently sabotage the paper with the publisher on some information unknown to you, I wonder what academic freedom means in that field. So you have to show perseverance to get your points out. 

Going forward, are you planning on any cross disciplinary collaborations? 

Yes, I’m working with historians on laying bare the biases in the construction of Standard Canadian English. Indigenous colleagues offer important perspectives on the colonial gaze that would otherwise be difficult to come by. Here, a lot of decolonization work is necessary, but also fascist bias is not unknown in Canadian English. One of the main researchers in the 1960s was a self-confessed Canadian fascist, as we just found out recently. Historians understand that connection between linguistics and anthropology perfectly well. I hope that we linguists will catch up in that regard and don’t so blindly trust our data — which, after all, can never be neutral. Interpret, for instance, the data in a framework handed down to you and you inadvertently phase out some – often minority – perspectives. Linguistics, as a discipline, after all is – via its philological roots – a by-product of nationalism. We ought to face the legacies we haven’t dehegemonized yet. Language is socially malleable. And we, as linguists, have a role in that process, whether we like it or not. Big data, in a way, requires us to rethink some of the presuppositions we have never thought much about. What is language, what’s a dialect? It’s no solution to scoff at the question or consider it outside our field of study: what poor, narrow field would that be? It’s something socially very important to speakers and we have, as has been explored for a long time now, to pay attention.

If you had to choose a different career path, what would you be interested in pursuing?

Historian would be an obvious answer! Or maybe first mate on a two-master to far-away climes…

“Eberhard Kranzmayer’s Deutschtum: on the Austrian dialectologist’s pan-German frame of reference” has been published in the fall of 2023. If you are interested, please access it here. 

(Dollinger, Stefan. 2023. Eberhard Kranzmayer’s Deutschtum: on the Austrian dialectologist’s pan-German frame of reference. Journal of Austrian Studies 56(3): 62-89)