• Digital Humanities and Cultural Heritage
• Publicationsclose
• Research dataclose
• Research softwareclose
• Open Access close
• FI close
• English close

Interdisciplinary research in the fields of forestry and sustainability studies often encounters seemingly incompatible ontological assumptions deriving from natural and social sciences. The perceived incompatibilities might emerge from the epistemological and ontological claims of the theories or models directly employed in the interdisciplinary collaboration, or they might be created by other epistemological and ontological assumptions that these interdisciplinary researchers find no reason to question. In this paper we discuss the benefits and risks of two possible approaches, Popperian optimism and Kuhnian pessimism, to interdisciplinary knowledge integration where epistemological and ontological differences between the sciences involved can be expected. Peer reviewed

International audience

The task of bilingual dictionary induction (BDI) is commonly used for intrinsic evaluation of cross-lingual word embeddings. The largest dataset for BDI was generated automatically, so its quality is dubious. We study the composition and quality of the test sets for five diverse languages from this dataset, with concerning findings: (1) a quarter of the data consists of proper nouns, which can be hardly indicative of BDI performance, and (2) there are pervasive gaps in the gold-standard targets. These issues appear to affect the ranking between cross-lingual embedding systems on individual languages, and the overall degree to which the systems differ in performance. With proper nouns removed from the data, the margin between the top two systems included in the study grows from 3.4% to 17.2%. Manual verification of the predictions, on the other hand, reveals that gaps in the gold standard targets artificially inflate the margin between the two systems on English to Bulgarian BDI from 0.1% to 6.7%. We thus suggest that future research either avoids drawing conclusions from quantitative results on this BDI dataset, or accompanies such evaluation with rigorous error analysis. Accepted at EMNLP 2019

Polynczuk-Alenius's article contributes to a better understanding of the racist moment in Poland that began in the aftermath of the 'refugee crisis' in 2015. It does so by zooming in on Christian far-right discourse and reconstructing a cognitive map of the social world manufactured therein. To this end, it analyses the blog of the former Catholic priest Jacek Miedlar, now a far-right activist and one of the leaders of the anti-refugee movement. In doing so, the article relies on two compatible bodies of research that have rarely been used together. Theoretically, the article approaches Christian far-right discourse as an articulation of the paranoid style and concentrates on its conspiratorial aspect. Analytically, it uses the fourfold model of authoritarian communication developed by the Frankfurt School to dissect systematically the conspiratorial tale expounded on Miedlar's blog. Accordingly, the empirical analysis of 116 blog posts treats the following themes: 1) the discontent diagnosed by Miedlar (anti-Polonism, epitomized by the suppression of nationalist and Christian values in favour of European universalism); 2) the alleged operators of anti-Polonism (the Jewish-orchestrated conspiracy bent on dominating the world and its puppets); 3) the movement that will rise up against this cabal (namely, the Polish Catholic nationalists armed with conservative values); and 4) the leader of the struggle (Miedlar himself as a Christ-like martyr figure). The article concludes that the anti-Muslim discourse, premised on an appeal to racist sentiments, served as a gateway into the conspiratorial, deeply antisemitic world-view of the Christian far-right milieu. In Poland, as elsewhere, such a world-view, stored and transmitted through the fringe far-right discourse, usually seems to gain traction in wider society during times of crisis. Peer reviewed

This paper discusses and analyses the memor­ial complex of Tsitsernakaberd in Yerevan as an architectural and symbolic entity in relation to Armenian national identity in the aftermath of the Armenian genocide of 1915. How does this Soviet-era structure fulfil its role as a genocide memorial today, including its function as a forced substitute for the hundreds of holy places and the culture and life connected with them? On the one hand, this is only a small inquiry into the function of one building complex. Yet on the other hand, the topic is more essential than perhaps anything in history: the genocide memorial crystallises a set of profound questions, serious problems and agonising processes. An entire national existence can be crushed in a genocide and subsequently debased through its denial, resulting in existential problems such as, on the one hand, a pressure of assimilation for the diaspora, and on the other, severe socio-economic and geopolitical-military crises in present-day Armenia.

In our submission to the SIGMORPHON 2022 Shared Task on Morpheme Segmentation, we study whether an unsupervised morphological segmentation method, Morfessor, can help in a supervised setting. Previous research has shown the effectiveness of the approach in semisupervised settings with small amounts of labeled data. The current tasks vary in data size: the amount of word-level annotated training data is much larger, but the amount of sentencelevel annotated training data remains small. Our approach is to pre-segment the input data for a neural sequence-to-sequence model with the unsupervised method. As the unsupervised method can be trained with raw text data, we use Wikipedia to increase the amount of training data. In addition, we train multilingual models for the sentence-level task. The results for the Morfessor-enriched features are mixed, showing benefit for all three sentencelevel tasks but only some of the word-level tasks. The multilingual training yields considerable improvements over the monolingual sentence-level models, but it negates the effect of the enriched features. Peer reviewed

The article analyses the histories of Catholic material culture in revolutionary Cuba. It illustrates and discusses how the Cuban Church and individual Catholics have navigated the revolutionary ev...

This article explores the interrelatedness of societal changes and changes in language practices. By using a combination of corpus linguistic and socio-pragmatic methods, we track diachronic changes in word patterns and interpret findings in the framework of democratization. The data comes from a small and representative corpus of British English (ARCHER-3.1) and from three "big data" sets (Google Books, British Library Newspapers and The Economist). We suggest that data triangulation, including sociohistorical contextualization, allows us to conclude that especially from the mid-nineteenth century onwards words signaling social status and referring to individuals have decreased and from the first decades of the twentieth century onwards words referring to collectivities of people have increased. (C) 2020 The Authors. Published by Elsevier Ltd. Peer reviewed