Еще раз о формах показателя множественного числа в тюркских языках

Introduction. The article continues the discussion of isogloss types and their relevance for the Proto-Turkic reconstruction and reconstruction of the intermediate nodes of the Turkic family tree. Goals. The paper makes another attempt to reconstruct the morphophonological appearance of some affixes for intermediate languages-ancestors of the standard Turkic group (Oguz, ‘Kyrgyz’, Altai, Karluk, Toba, Kypchak). The study draws into consideration not only the plural affix *-lar, but in general inflectional and derivational affixes starting with *-l. Materials and Methods. Methods of stepwise reconstruction are used simultaneously with morphophonological methods of identifying classes of positions and distribution of classes of allomorphs. Field records of dialects, dialectological publications, both modern ones and those of the 19th century, as well as written monuments were used as research material. Results. Both modern field data and classical sources, with the correct application of the methods of stepwise reconstruction, point that affixal *-l has no alternants in proto-Oghuz, proto-Karluk and proto-Qypchaq. All instances of alternation in modern idioms like dialectal Bashkir, dialectal Kazakh, ‘Qyrghyz’ languages, Yakut-Dolghan and Toba languages are to be classified as recent areal innovation. This is deduced due to the nature of morphophonological rules in these languages — neither is applyable for the proto-Common-Turkic stem auslaut, but instead is limited to forms that are specific to each separate group in question.

The verified CakeML compiler backend

The CakeML compiler is, to the best of our knowledge, the most realistic verified compiler for a functional programming language to date. The architecture of the compiler, a sequence of intermediate languages through which high-level features are compiled away incrementally, enables verification of each compilation pass at an appropriate level of semantic detail. Parts of the compiler’s implementation resemble mainstream (unverified) compilers for strict functional languages, and it supports several important features and optimisations. These include efficient curried multi-argument functions, configurable data representations, efficient exceptions, register allocation, and more. The compiler produces machine code for five architectures: x86-64, ARMv6, ARMv8, MIPS-64, and RISC-V. The generated machine code contains the verified runtime system which includes a verified generational copying garbage collector and a verified arbitrary precision arithmetic (bignum) library. In this paper, we present the overall design of the compiler backend, including its 12 intermediate languages. We explain how the semantics and proofs fit together and provide detail on how the compiler has been bootstrapped inside the logic of a theorem prover. The entire development has been carried out within the HOL4 theorem prover.

Large-scale Vocabulary Surveys as a Tool for Linguistic Stratigraphy: A California Case Study

In lieu of an abstract, here is a brief excerpt:This paper presents the results of a comprehensive search for lookalikes among words for plants and animals in California languages. Words in this domain are typically more prone to borrowing than basic vocabulary, especially when speakers of a language move and encounter different species. A survey of such vocabulary is especially suited to identifying and highlighting old language contact. Since a language may be spoken far away from where its ancestor was once in contact with another language, and since words may spread far from their source through intermediate languages, this study does not exclude any languages in the area from being ultimately interconnected though old contact events.