#  Две старые машины с переполненными багажниками - Two talks on the Russian language from the top of the world featuring 

 



####  calendar\_today Date and Time 

 **October 30, 2017** 

 04:00PM - 06:00PM EDT 

####  pin\_drop Location 

 **Barker 373 (Klasson Room)**  



 

 



 

 Tore Nesset on why we say две старые машины ‘two old cars’

 Laura Janda on why our Russian language textbooks are like cars with overstuffed trunks

 \* \* \*

 Why we say две старые машины Tore Nesset, University of Tromsø, The Arctic University of Norway

 In this talk, I show that the Russian language is in the process of changing. First, based on extensive evidence from electronic corpora it is demonstrated that the rivalry between the accusative and locative cases in expressions like v dvadcatye gody vs. v dvadcatyx godax ‘in the twenties’ has changed radically over the last two centuries, and secondly a similarly dramatic change is documented for the rivalry between the accusative and genitive cases in examples like dve starye mašiny vs. dve staryx mašiny ‘two old books’. On the basis of these two case studies, I discuss the general question as to how far corpus data can take us in analyzing language change.

 Why our Russian language textbooks are like cars with overstuffed trunks

 Laura A. Janda, University of Tromsø, The Arctic University of Norway and Francis M. Tyers, National Research University Higher School of Economics

 Learning foreign languages typically involves memorizing tables of forms, called “paradigms”, but does it have to be that way? And, more importantly, is that the most efficient way to learn a language? We show that most of the forms in paradigms are rarely or never encountered in authentic language use. We have conducted a computational learning experiment comparing the prediction of Russian word forms using two types of training data: full paradigms vs. single word forms for Russian nouns, verbs, and adjectives. We find that learning is most accurate and reliable when individual high frequency forms are used than when full paradigms are learned. This is probably because training on full paradigms overpopulates the search domain with highly unlikely word forms.



 

 



 

 

 Share on:- [     Facebook ](#)
- [     Twitter ](#)
- [     Linkedin ](#)
 


 Save: [ Add to calendar calendar\_today ](https://slavic.fas.harvard.edu/node/1076436/event-feed.ics)  Copy link link