Modelling the Dynamics of Language Change: Logistic Regression, Piotrowski’s Law, and a Handful of Examples in Polish

ABSTRACT

The study discusses modelling diachronic processes by logistic regression. The phenomenon of nonlinear changes in language was first observed by Raimund Piotrowski (hence labelled as Piotrowski’s law), even if actual linguistic evidence often speaks against using the notion of a ‘law’ in this context. In our study, we apply logistic regression models to changes which occurred between 15^th and 18^th century in the Polish language. The attested course of the majority of these changes closely follow the expected values, which proves that the language change might indeed resemble a nonlinear phase change scenario. We also extend the original Piotrowski’s approach by proposing polynomial logistic regression for these cases which can hardly be described by its standard version. Also, we propose to consider individual language change cases jointly, in order to inspect their possible collinearity or, more likely, their different dynamics in the function of time. Last but not least, we evaluate our results by testing the influence of the subcorpus size on the model’s goodness-of-fit.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1. To be precise, phonologically this is a change of the phoneme /i/ > /e/ following a palatal consonant. Since palatalization is orthographically marked by the letter i following a consonant, the process is reflected by substitution of a string of letters ir with ier.

2. Due to copyright restrictions, the corpus could not be made publicly available. However, we post all the derivative datasets used in this study, as well as the full set of the results, followed by the code needed to replicate the tests, on our GitHub repository: https://github.com/computationalstylistics/PL_language_change

3. All the datasets used in this study, together with the final plots and an R script to replicate the results, are posted on our GitHub repository: https://github.com/computationalstylistics/PL_language_change.

4. Note that in Polish, comparative and superlative share the same suffix, while the latter differs from former by the prefix naj-.

5. To be precise, apart from the above mentioned there are also markers of dual: -swa and -chwa. However, since dual went out of use in the early 16^th century, we will gloss over these morphemes.

6. In our scrutiny, we did not take into account variants cirzpieć, pirzwej etc. due to their scarcity in the dataset.

7. Including the plural forms -bychmy > -byśmy.

Additional information

Funding

This research was conducted as a result of a project supported by Poland’s National Science Centre [project number UMO-2013/11/B/HS2/02795]. This work was supported by the Narodowe Centrum Nauki