Why do you talk like that? �Minority languages and politics

Ian Stewart

LING 3810

26 March 2019

https://www.planetware.com/map/spain-language-areas-of-spain-map-e-e5.htm

https://www.planetware.com/map/spain-language-areas-of-spain-map-e-e5.htm

https://gl.wikipedia.org/w/index.php?title=File:Linguistic_map_Southwestern_Europe.gif

http://www.languagesgulper.com/eng/Northamer.html

https://slate.com/culture/2014/05/language-map-whats-the-most-popular-language-in-your-state.html

Overview

• Minority languages in everyday life
• Definitions
• Legal rights
• Technological considerations
• Political identity
• Revitalization efforts

Minority languages in everyday life

What counts as a minority language?

• Legal definition: language that is not the “numerical majority.”
• European Charter for Regional or Minority Languages
• “traditionally used within a given territory of a State by nationals of that State who form a group numerically smaller than the rest of the State's population; and different from the official language(s) of that State”
• Usually the result of immigration or marginalization
• Linguistic definition: linguistically distant or socially marked (Kloss 1967)
• Abstand (“language”) = differences in production/understanding
• Ausbau (“building-out”) = differences in development/standards
• Subcategories:
• Endangered language = at-risk language whose speakers are dying or forgetting faster than learning
• Heritage language = language learned at home, usually underdeveloped

What counts as a minority language?

• Legal definition: language that is not the “numerical majority.”
• European Charter for Regional or Minority Languages
• “traditionally used within a given territory of a State by nationals of that State who form a group numerically smaller than the rest of the State's population; and different from the official language(s) of that State”
• Usually the result of immigration or marginalization
• Linguistic definition: linguistically distant or socially marked (Kloss 1967)
• Abstand (“language”) = differences in production/understanding
• Ausbau (“building-out”) = differences in development/standards
• Subcategories:
• Endangered language = at-risk language whose speakers are dying or forgetting faster than learning
• Heritage language = language learned at home, usually underdeveloped

Fig. 3, Kloss (1967)

Official recognition

http://www.un.org/en/events/motherlanguageday/

European Charter for Regional or Minority Languages�(dark green=sign+ratify, light green=sign, white=none)

Legal rights for minority languages

• Need for minority language accommodations at multiple levels:
• Official signs
• Translations of documents, court proceedings
• Interpreters for courts, hospitals
• Teachers
• Longer history of multilingualism may provide more federal rights.
• Canada’s Official Languages Act gives “preferred” status to French and English, ensuring that neither will be treated as minority
• Court protection, work protection, bilingual parliament proceedings
• Ireland treats Irish as national language, even though it is minority

• More often, decisions are made at local level.
• No national language in U.S.; majority of states have English as official language
• Arizona: (English) official language; (Spanish/Native) ballot translation, right to interpreter at polls
• California: (English) official language; (variety) DMV information, (bilingual) education
• New Mexico: (Spanish) education, information about new laws, ballot information

Accommodating to immigrants in the U.S.

• Court interpreters have extremely difficult job: must translate unambiguous details under stressful situation; legal language is a different “domain” than everyday language.�
• Default assumption: most immigrants speak Spanish.�
• How can legal system accommodate long tail of minority languages?
• Hire more interpreters?
• Phone calls?
• Intermediary language?

https://www.nytimes.com/2019/03/19/us/translators-border-wall-immigration.html

Case study: Linguistic landscapes

• Linguistic landscape - “visibility and salience of languages on public and commercial signs in a given territory or region” (Landry and Bourhis 1997).
• The presence of signs in everyday situations reminds us of the diversity of languages in our environment, as well as power relations between languages.
• Example: strong language policy protects Basque (Spain) speakers and encourages more Basque on signs as compared to Frisian (Netherlands) speakers who have less protection (Cenoz and Gorter 2006).

Case study: Linguistic landscapes

• When demand for minority languages outweighs supply...

https://www.huffingtonpost.com/2013/02/27/10-funny-translations-spanish-english_n_2774588.html

Technological considerations

• Common Natural Language Processing applications: machine translation, speech recognition, search.
• Most NLP applications require large amount of data to perform well.
• Minority languages often have fewer resources, less support from existing systems.
• Solutions:

Technological considerations

• Common Natural Language Processing applications: machine translation, speech recognition, search.
• Most NLP applications require large amount of data to perform well.
• Minority languages often have fewer resources, less support from existing systems.
• Solutions:
• Collect more data! Problem: expensive, difficult to find annotators.
• Transfer from existing data, e.g. Spanish to Catalan. Problem: sometimes no similar languages exist (Basque).
• Assume similar structures across languages and transfer anyway. Problem: not enough data to infer structure.
• Political consequences: major populations are excluded from written/online discussions.

Smith et al. (2017)

Translation fails: Gaelic

More resources = better?

More resources = better?

Case study: machine translation in Haiti

Low-resource language issues

Political identity

Political identity

• Political debate thrives on concrete cultural symbols, including language.
• Politicians signal loyalty, gain attention through cultural cues.
• Regional politics are guided by demographics and socially constructed identity.
• U.S.: English/Spanish
• Spain: Spanish/Catalan/Galician/Basque
• Scotland: English/Scots
• Social identity theory (Tajfel and Turner 1979)
• Personal identity comes from group affiliation, which defines social expectations.
• Positive stance toward in-group, negative stance toward out-group.
• Interpersonal conflict is often driven by opposing relationships between groups, rather than individuals.
• Indexicality (Eckert 2008)
• Social/linguistic stylistic choices help individuals construct dynamic identities in conversation and interaction.
• Example: use of minority language among friends.

Political identity: U.S. immigrants

http://time.com/5416077/colorado-grocery-store-viral-video-harassment/

https://www.cnn.com/2018/05/17/us/new-york-man-restaurant-ice-threat/index.html

Political identity: independence

• Secession is the most extreme political act (from nation-state perspective), and therefore the one most likely to attract a wide variety of attention.
• Push for independence driven partly by identity, expressed through language.�
• In 2014, Scotland held a referendum to secede from the UK.
• Massively divided the voting population due to economic and cultural considerations.

https://www.itv.com/news/2016-07-19/trident-renewal-intensifies-calls-for-scottish-independence-vote/

https://www.thetimes.co.uk/article/all-under-one-banner-march-in-glasgow-piles-on-pressure-for-second-scottish-independence-referendum-r6mxb537w

Overview

• Scotland referendum
• 45% for independence
• Code-mixing (word)
• 30% Scots understanding

• Catalonia referendum
• 92% for independence
• Code-switching (sentence)
• 80% Catalan understanding

Aye or naw, whit dae ye hink? Scottish independence and linguistic identity on social media�(Shoemark et al., 2017)

Sí o no, ¿què penses? Catalonian Independence and Linguistic Identity on Social Media�(Stewart, Pinter, Eisenstein 2018)

26

Politics and local languages

• Language variation often reveals aspects of social identity, such as race and gender.
• Local languages have special connection to regional and political identity.
• The role of language in political identity expression is magnified during controversy.

How does local language use correlate with political expression on social media?

27

Study objectives

RQ1: Is a speaker’s stance on independence strongly associated with their Catalan usage?��RQ2a: Does Catalan usage vary depending on whether the discussion topic is related to the referendum?�RQ2b: Does Catalan usage vary depending on the imagined audience?

28

29

RQ1: Do separatists tweet more with local language?

d=0.2523 (p<0.001)

d=0.00555 (p<0.01)

30

RQ2a: Is local language more likely in referendum discourse?

p<0.01

p=0.079

p<0.01

p=0.111

31

Yes

No

Conclusions

• Pro-independence tweeters more likely to use Catalan than anti-independence tweeters.
• Tweeters more likely to use Catalan for broad audience than for narrow audience.
• Local language less likely to be useful with replies (mostly monolingual Spanish speakers).
• More extreme language patterns in Catalonia versus Scotland.
• Linguistic identity could be more pronounced with code-switching than with code-mixing.
• Future work: political expression in societies with more evenly distributed languages (Belgium).

32

Revitalization

Language extinction

• Up to 50% of world’s languages could be extinct by 2100 (Krauss 1992)
• Determining language vitality (UNESCO 2003):
• Intergenerational transmission
• Number of speakers
• Proportion of speakers
• Use in existing contexts
• Spread to new contexts
• Availability of educational material
• Institutional policies
• Community attitudes
• Documentation quality

https://anthropology.net/2007/09/19/5-hotspots-where-languages-are-becoming-extinct/

Revitalization

• Native U.S. languages were suppressed from 18th century onward, resulting in failure to transmit language between generations.
• People in postcolonial societies see revitalization as a way to support (precolonial) minority languages with cultural significance.
• Methods for revitalization
• Education
• Improved access in everyday life
• Increased cultural interest
• Educators, activists and policy-makers have vested interest in maintaining traditional culture for the sake of local population and non-local tourists.
• Political benefits: strengthening national identity (Ireland), improving status of native population (New Zealand), preserving traditions for future generations (United States).

https://www.languageconservancy.org/news-events/2014/06/05/revitalizing-endangered-language-part-6/

Case study: Maori

• Native language of New Zealand
• Typologically unusual: VSO, few vowel/consonants, dual pronouns.
• Post-colonization (1980s): < 20% (~10,000s) of native population spoke Maori
• Rich oral tradition among Maori led to language nest (kohanga reo)
• Founded in 1982 by Maori educators
• Goal: total immersion in Maori language and culture for young students (0-6)
• Expanded to several hundred operations, thousands of children
• Challenge: imperfect transmission (phonology, morphology, etc.), training teachers (Reedy 2000), maintaining cultural context (Hohepa et al. 2010).
• 2013 Census: ~ 125,000 speakers

Commericalization of Maori

Strategies for successful revitalization

• Community-driven > top-down (Grenoble and Whaley 2006).
• Ensure that native speakers, communities have maximum authority in preservation.�
• Immersion is good goal, especially when it maximizes exposure to culture, but bilingual speakers are more realistic (Peter 2003).�
• Ensure long “lifespan” of language by providing community with opportunities to practice language in real-world events, e.g. ceremonies (Hermes 2007).

https://www.languageconservancy.org/programs/indigenous-language-program-support/

Recap

• Minority languages in everyday life
• Definitions
• Legal rights
• Technological considerations
• Political identity
• Revitalization

Recap

• Minority languages in everyday life
• Definitions
• Legal rights
• Technological considerations
• Political identity
• Revitalization

People don’t speak languages in a vacuum! The status of the language that we speak often has important political consequences with respect to government and society.

Sources

Cenoz, J., & Gorter, D. (2006). Linguistic landscape and minority languages. International Journal of Multilingualism, 3(1), 67-80.�Eckert, P. (2008). Variation and the indexical field. Journal of sociolinguistics, 12(4), 453-476.�Esses, V. M., & Gardner, R. C. (1996). Multiculturalism in Canada: Context and current status. Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement, 28(3), 145.�Grenoble, L. A., & Whaley, L. J. (2006). Saving languages: An introduction to language revitalization. Cambridge University Press. �Hermes, M. (2007). Moving toward the language: Reflections on teaching in an Indigenous-immersion school. Journal of American Indian Education, 46(3), 54-71.�Hohepa, M., Smith, L. T., & McNaughton, S. (1992). Te Kohanga Reo hei Tikanga ako I te Reo Maori: Te Kohanga Reo as a context for language learning. Educational Psychology, 12(3-4), 333-346.�Kloss, H. (1967). 'Abstand languages' and'ausbau languages'. Anthropological linguistics, 29-41.�Krauss, M. (1992). The world’s languages in crisis. Language, 68(1), 4-10.�Landry, R., & Bourhis, R. Y. (1997). Linguistic landscape and ethnolinguistic vitality: An empirical study. Journal of language and social psychology, 16(1), 23-49.�Lewis, W. (2010, May). Haitian Creole: How to Build and Ship an MT Engine from Scratch in 4 days, 17 hours, & 30 minutes. In 14th Annual conference of the European Association for machine translation.�McQueeney, K. B. (2001). Language Policy and Identity Politics in the United States. Contemporary Sociology, 30(6), 623.�Peter, L. (2003). Assessing the Impact of Total Immersion on Cherokee Language Revitalization: A Culturally Responsive, Participatory Approach.�Reedy, T. (2000). Te Reo Māori: The past 20 years and looking forward. Oceanic Linguistics, 157-169.�Shoemark, P. Debnil Sur, Luke Shrimpton, Iain Murray, and Sharon Goldwater. 2017. Aye or naw, whit dae ye hink? Scottish independence and linguistic identity on social media. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics (Vol. 1, pp. 1239-1248).�Smith, S. L., Turban, D. H., Hamblin, S., & Hammerla, N. Y. (2017). Offline bilingual word vectors, orthogonal transformations and the inverted softmax. ICLR 2017.�Stewart, I., Pinter, Y., & Eisenstein, J. (2018). Si O No, Que Penses? Catalonian Independence and Linguistic Identity on Social Media. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) (Vol. 2, pp. 136-141).�Tajfel, H., Turner, J. C., Austin, W. G., & Worchel, S. (1979). An integrative theory of intergroup conflict. Organizational identity: A reader, 56-65.