Due to the Corona virus education methods or examination can deviate. For the latest news please check the course page in Brightspace.

Prospectus

nl en

Topics in Linguistics E: Bayesian Computational Methods in Historical Linguistics

Course
2017-2018

This is an extra Topics course, only open to students of the ResMA Linguistics.

Description

The ‘Topics in Linguistics’ courses offer up to date courses on linguistic research by LUCL staff. The actual content of these courses vary.

This year’s ‘Topics in Linguistics E’ course takes a look at the recent trend of Bayesian phylogenetics in historical linguistics, where big linguistic datasets, such as collections of word lists, are used in computer methods to re-construct family trees. The course will provide the background necessary for critically evaluating studies employing such methods. To achieve this, the course will be split between lectures on the theoretical foundations of Bayesian inference using computers and critical reading of literature articles from the topic.

Course objectives

  • Students will be aware of opportunities and limits of Bayesian models.
    • Students will have a good grasp of the computer modelling sociolect. This includes concepts such as “Markov chain Monte Carlo method”, “robustness”, “garbage-in/garbage-out”, “likelihood”, “marginal posterior probability”, “uninformative prior”, “pre-processing” etc.
    • Students will be able to assess the value of results from Bayesian phylogenetic modelling literature.

Timetable

2 h weekly Linguistics research

Mode of instruction

  • Seminar

Course Load

Total course load: 140 h (5EC)

  • Amount of lectures: 2 h per week × 7 weeks = 14 h

  • Preparation lectures/Compulsory Literature: 7 h per week × 7 weeks = 49 h

  • Preparation Assignment 2: 25 h

  • Assignment 1 (Article Assessment): 30 h

  • Assignment 2 (Written Examination): 2h

  • Assignment 3 (Peer Review): 20 h

Assessment method

  • Written examination with open questions. (40%)

  • Paper: An assessment of one article from a selection taken from the Bayesian inference literature, in the style of a peer review. (40%)

  • Short Paper: Reviewing the article assessments written by a fellow student. (20%)

Weighing

The final mark for the course is established by determining the weighted average according to the percentages given above.

Resit

Resit is available for all parts of the assignment that got marks of 5.5 or less.

Exam review

How and when an exam review will take place will be disclosed together with the publication of the exam results at the latest. If a student requests a review within 30 days after publication of the exam results, an exam review will have to be organized.

Blackboard

Blackboard will be used for:

  • Course information and course documents

  • Reading list

Reading list

The reading list consist of a selection of articles from the Bayesian phylogenetics literature. The reading list will be made available through Blackboard.

Registration

Enrolment through uSis for the course and the examination or paper is mandatory.
General information about uSis is available in English and Dutch

Contact

Dr. G.A. Kaiping

Education Administration Office van Wijkplaats: osz-oa-wijkplaats@hum.leidenuniv.nl

Coordinator of Studies: Else van Dijk

Remarks

Students should be familiar with some basic concepts from historical linguistics, such as “cognate”, “systematic sound change” and “reconstructed proto-languages”.
A passing familiarity with scientific terms related to the contents of the course will be helpful: Concepts such as “logarithm”, “conditional probability”, and “mutation” will be explained shortly, but then taken for granted.