Søren Wichmann


pdf bib
Finding Concept-specific Biases in Form–Meaning Associations
Tiago Pimentel | Brian Roark | Søren Wichmann | Ryan Cotterell | Damián Blasi
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

This work presents an information-theoretic operationalisation of cross-linguistic non-arbitrariness. It is not a new idea that there are small, cross-linguistic associations between the forms and meanings of words. For instance, it has been claimed (Blasi et al., 2016) that the word for “tongue” is more likely than chance to contain the phone [l]. By controlling for the influence of language family and geographic proximity within a very large concept-aligned, cross-lingual lexicon, we extend methods previously used to detect within language non-arbitrariness (Pimentel et al., 2019) to measure cross-linguistic associations. We find that there is a significant effect of non-arbitrariness, but it is unsurprisingly small (less than 0.5% on average according to our information-theoretic estimate). We also provide a concept-level analysis which shows that a quarter of the concepts considered in our work exhibit a significant level of cross-linguistic non-arbitrariness. In sum, the paper provides new methods to detect cross-linguistic associations at scale, and confirms their effects are minor.


pdf bib
The DReaM Corpus: A Multilingual Annotated Corpus of Grammars for the World’s Languages
Shafqat Mumtaz Virk | Harald Hammarström | Markus Forsberg | Søren Wichmann
Proceedings of the Twelfth Language Resources and Evaluation Conference

There exist as many as 7000 natural languages in the world, and a huge number of documents describing those languages have been produced over the years. Most of those documents are in paper format. Any attempts to use modern computational techniques and tools to process those documents will require them to be digitized first. In this paper, we report a multilingual digitized version of thousands of such documents searchable through some well-established corpus infrastructures. The corpus is annotated with various meta, word, and text level attributes to make searching and analysis easier and more useful.

pdf bib
From Linguistic Descriptions to Language Profiles
Shafqat Mumtaz Virk | Harald Hammarström | Lars Borin | Markus Forsberg | Søren Wichmann
Proceedings of the 7th Workshop on Linked Data in Linguistics (LDL-2020)

Language catalogues and typological databases are two important types of resources containing different types of knowledge about the world’s natural languages. The former provide metadata such as number of speakers, location (in prose descriptions and/or GPS coordinates), language code, literacy, etc., while the latter contain information about a set of structural and functional attributes of languages. Given that both types of resources are developed and later maintained manually, there are practical limits as to the number of languages and the number of features that can be surveyed. We introduce the concept of a language profile, which is intended to be a structured representation of various types of knowledge about a natural language extracted semi-automatically from descriptive documents and stored at a central location. It has three major parts: (1) an introductory; (2) an attributive; and (3) a reference part, each containing different types of knowledge about a given natural language. As a case study, we develop and present a language profile of an example language. At this stage, a language profile is an independent entity, but in the future it is envisioned to become part of a network of language profiles connected to each other via various types of relations. Such a representation is expected to be suitable both for humans and machines to read and process for further deeper linguistic analyses and/or comparisons.


pdf bib
How to Distinguish Languages and Dialects
Søren Wichmann
Computational Linguistics, Volume 45, Issue 4 - December 2019

The terms “language” and “dialect” are ingrained, but linguists nevertheless tend to agree that it is impossible to apply a non-arbitrary distinction such that two speech varieties can be identified as either distinct languages or two dialects of one and the same language. A database of lexical information for more than 7,500 speech varieties, however, unveils a strong tendency for linguistic distances to be bimodally distributed. For a given language group the linguistic distances pertaining to either cluster can be teased apart, identifying a mixture of normal distributions within the data and then separating them fitting curves and finding the point where they cross. The thresholds identified are remarkably consistent across data sets, qualifying their mean as a universal criterion for distinguishing between language and dialect pairs. The mean of the thresholds identified translates into a temporal distance of around one to one-and-a-half millennia (1,075–1,635 years).


pdf bib
Towards identifying the optimal datasize for lexically-based Bayesian inference of linguistic phylogenies
Taraka Rama | Søren Wichmann
Proceedings of the 27th International Conference on Computational Linguistics

Bayesian linguistic phylogenies are standardly based on cognate matrices for words referring to a fix set of meanings—typically around 100-200. To this day there has not been any empirical investigation into which datasize is optimal. Here we determine, across a set of language families, the optimal number of meanings required for the best performance in Bayesian phylogenetic inference. We rank meanings by stability, infer phylogenetic trees using first the most stable meaning, then the two most stable meanings, and so on, computing the quartet distance of the resulting tree to the tree proposed by language family experts at each step of datasize increase. When a gold standard tree is not available we propose to instead compute the quartet distance between the tree based on the n-most stable meaning and the one based on the n + 1-most stable meanings, increasing n from 1 to N − 1, where N is the total number of meanings. The assumption here is that the value of n for which the quartet distance begins to stabilize is also the value at which the quality of the tree ceases to improve. We show that this assumption is borne out. The results of the two methods vary across families, and the optimal number of meanings appears to correlate with the number of languages under consideration.