 Let's close out the week with a #geoweirdness thread Today we'll explore the wonderful world of linguistic maps and atlases. These thematic maps plot diatopic (dia = across, topy = place) distribution of linguistic features such as vocabulary, grammar usage and, most famously, pronunciation. Let’s get straight to it with a challenge: Below are three maps. For today’s thread, we’re interested in the info conveyed in just two of them. Can you work out which? 1/ 17 comments
3/ Now for a quick crash course in linguistics: There are many branches, but for today, it’ll be helpful to understand the following in a bit more detail: • Dialectology - the study of subsets of language (dialects) ex: Scouse English, Bavarian German • Phonetics - the study of sounds of speech - cat vs chat, ik vs ich • Sociolinguistics - the study of how social factors (i.e. age, class) impact speech These, of course, are not official definitions of these terms! Each of the maps in a linguistic atlas will explore the distribution of a different variant (a sound, a word, a phrase) using a sequence of points on a map An isogloss is like a variant boundary. It’s a line drawn on the map to show where a linguistic variant is found. https://en.wikipedia.org/wiki/Isogloss When several of these lines intersect, you can consider it a dialect boundary. I.e. all these features together are what give you a specific dialect like , for example. 5/ Time for some history. There have been thematic maps since at least 350 BC, but many linguists agree the first significant linguistic map was Gottfried Hensel’s famous 1741 map showing linguistic variation across Europe in the Lord’s Prayer. https://en.wikipedia.org/wiki/Synopsis_Universae_Philologiae 6/ Fast forward a century or two, and the atlas experiences its golden age. The first explicitly named linguistic atlas was Julius Klabroth’s 1823 “Asia Polyglotta: Sprachatlas”, followed by Adriano Balbi’s 1826 “Atlas ethnographique du Globe”. These atlases differ from modern ones because the data collected is mostly in comparative tables. The table below illustrates this point well, whilst the map magnificently shows isoglosses. 7/ The end of the 19th century saw the emergence of the two godfathers of language cartography, Georg Wenker, of Germany, and Jules Gilliéron, of Switzerland. They were the first to place maps at the absolute front and centre of their atlases, and developed many of the key design and data collection principles used today. Wenker's "Sprachatlas" lives on today in digital form at the University of Marburg: https://www.uni-marburg.de/en/fb09/dsa/projects/completed-projects/diwa 8/ Broadly speaking, Wenker is famous for using "indirect data collection" techniques. His first major work involved sending a questionnaire of 40 standardised sentences to school by mail. It was translated into dialects by pupils or teachers in about 40,000 locations in Germany, and the collected data on variants was plotted by hand onto over 1,500 maps. https://de.wikipedia.org/wiki/Deutscher_Sprachatlas#Wenkers%C3%A4tze 9/ By contrast, Gilliéron is known for direct data collection. In his first publication, Petit Atlas Phonétique du Valais Roman (Sud du Rhone), 1880, Gilliéron gathered necessary data by hiking through the southern Rhone valley, manually transcribing phonetic differences between small villages in the region. This data was then plotted onto a series of maps. 10/ Gilliéron most famous project is the Atlas linguistique de la France (ALF), which was first published in 1902. He employed a field worker, Edmont, to travel to 638 locations in France to transcribe data and produce 1,900 maps. It was also the first atlas to layer in sociolinguistic data, such as on age i.e. “chez les jeunes” (among young people). This work is also now online: https://lig-tdcge.imag.fr/cartodialect5/#/ 11/ Inclusion of critical sociolinguistic variation such as ethnicity, gender and level of education became commonplace in atlas projects of the early 20th century, with notable works including the Linguistic Atlas of New England (LANE). This was one of the first attempts to map speech in the US 🇺🇸. Once again, the results, along with all the annotated notes, have been digitalised here: https://linguisticatlasproject.org/LANE 12/ The next major development in linguistic atlases was the introduction of computers to streamline and improve the data collection and cartographic processes. Early successful examples of this include the Computer Developed Linguistic Atlas of England (CLAE, Viereck and Ramisch). Below you can see some of the code used to map the linguistic item “anvil” and its variants, and what this looks like on a modernised, more detailed map. 13/ Most new research/effort is focused on digitalising existing databases to preserve them and enhance accessibility, but some cutting-edge work is also being done to improve on this work using more modern GIS, geospatial analysis, and cartographic methodology. Take the *Atlas linguistique de la France* The data points on the map are each labelled with Investigation points, which can then be remapped onto modern maps using coordinates to produce more accurate final projects. 15/ For our 🇺🇸 American and 🇨🇦 Canadian friends - you will enjoy the amazingly detailed North American English Dialects map 16/ And finally, because big is brilliant, let’s end this thread on the World Atlas of Language Structures. An incredible study of over 160 maps showing the geographic distribution of structural linguistic features such as word order, number of genders and more. You can see the complete list of linguistic items here and find the maps in the links. https://wals.info/chapter We’ll leave you with a screenshot of one of the maps: 17/ So there you have it, linguistic atlases! Pretty fascinating, don’t you think? Be honest… how long did you spend listening to that fable and browsing the maps? Thank you very much for reading (and sharing)! We have more threads on #geoweirdness in specific regions and on other topics available over on our blog:  @opencage I hate to be that guy, but the title of the map is inaccurate as it implies Portugal is part of Spain. |
Gregory's Server
2/ The map of Iberia "only" plots the different regional languages, it doesn’t convey anything regarding language characteristics.
So it is simply a “map of languages” rather than a “linguistic map”.
We’re interested in the other two, which are great examples of linguistic maps which plot lexical (words) and syntactic (sentence structure) variation across geographical space.
#geoweirdness