Assessment of linguistic complexity with constraint-based tools

This project is placed in the area of linguistic complexity studies. We address complexity of natural languages ​​from a non-absolute perspective and we propose a constraint-based computational model that allows us to calculate relative linguistic complexity.
It has been defended for a long time that linguistic complexity is invariant and that languages are not measurable in terms of complexity. However, the validity of this claim has rarely been subjected to systematic crosslinguistic investigation. This idea of equicomplexity, seen for decades as an unquestioned truism of Linguistics, has begun to be explicitly questioned in recent years. The interest on linguistic complexity has led researchers to challenge the equicomplexity dogma by addressing the study of complexity from different points of views.
Despite the efforts made, and although it is commonly accepted that languages ​​exhibit different levels of complexity, it is not easy to calculate exactly those differences and we still lacking an answer on linguistic complexity and an effective method to measure it. Part of that difficulty is due to the different ways of understanding complexity in natural languages. One of the most used typologies of complexity in Linguistics is the one that distinguish between absolute complexity, an objective property of the system (measured in terms of the number of parts of the system), and relative complexity, that takes into account the users of language and identifies complexity with difficulty/cost of processing, learning or acquisition.
Eventhough most of the works carried out on linguistic complexity ​​adopts an absolute perspective, many specialists are interested in analyzing relative complexity taking into account the process of second language learning​​ (L2). The main part of these works focuses exclusively on measuring the complexity of the target language, obviating the relevance that learner's mother tongue (L1) may have to determine relative complexity.
In this project, we argue that linguistic complexity can only be determined by jointly considering absolute and relative complexity; and that L1 is decisive when calculating the relative linguistic complexity of an L2 in terms of learning difficulty.
The general goals of this project are the following: to contribute to studies on linguistic complexity by demonstrating that languages ​​differ in relative complexity; to provide an objective method to calculate the relative complexity of natural languages.
In order to reach our goals, we propose the use of a computational tool to calculate the linguistic complexity in terms of learning difficulty: womb grammars. This model can be seen as an alternative to the methods that have been used to calculate linguistic complexity and it has the following advantages: 1) It allows to calculate relative complexity considering adults who learn an L2 (most of the works adopt an absolute perspective and are scarce those that address the complexity from the point of view of the user); 2) It calculates the complexity by taking into account a contrastive analysis of the L1 and L2 systems (most of the studies pay attention, almost exclusively, to the target language and the percentage of success shown by the learners); 3) Being a computational tool, it avoids the problem of the influence of external (non-linguistic) factors that can condition the learning process and provides objective and quantifiable data to calculate linguistic complexity.