Full-text resources of CEJSH and other databases are now available in the new Library of Science.
Visit https://bibliotekanauki.pl

Results found: 50

first rewind previous Page / 3 next fast forward last

Search results

Search:
in the keywords:  ETYMOLOGY
help Sort By:

help Limit search:
first rewind previous Page / 3 next fast forward last
EN
The aim of this minor contribution to Afro-Asiatic (Semito-Hamitic) etymology is to appraise in the light of recent results of the named field of research the achievements of F. von Calice, a pioneer of Egyptian etymology, whose magnum opus appeared exactly 70 years ago. The methods applied therein are those of comparative Afro-Asiatic linguistics: an attempt is made at evaluating each selected etymology according to the criteria of semantics and consonantal correspondences. As a result, the reader can hopefully measure the progress of this minor field in the past 70 years.
2
Content available remote

TOCHARIAN B PÄST/PEST 'AWAY'

100%
EN
According to J.H.W. Penney, there are three views on the difference between Tocharian B päst and pest 'away': (1) Pest is stressed and päst unstressed. (2) Päst - *pe-sth2-u and pest - *po-sth2-u. (3) -We are dealing with a vowel which began as something that could be written as 'e' but was not identical to the 'e' that survived intact, and which at the time of the reordering of the Tocharian B vowel system shifted to fill the slot left vacant by the move of earlier stressed 'ä' to 'a' -. The present writer believes that päst is the basic form, while pest arose by what he calls irregular sound change due to frequency.
3
Content available remote

LUSITANIEN 'REVE' - LATIN 'JOVI'

100%
EN
The aim of this contribution is to give on overview of the different conceptions regarding the etymology of Lusitanian word 'Reve'. In Lusitanian, the name of a god is attested nine times in the dative singular 'Reve'. There are different opinions on the etymology of this theonym. According to F. Villar, Lus. 'Reve' = Lat. 'Rivus'. K.T. Witczak believes that the god in question is an exact equivalent of Greek Zeus and Roman Iupiter (i.e. 'Reve' < *diu-ei or *dyeu-ei) and claims that in Lusitanian *d > r. But F. Vilar and R. Pederero enumerate many place names with d unaltered. According to the present author, the initial consonant in 'Reve' is to be accounted for by what he calls irregular sound change due to frequency.
EN
The author gives a typology of corrections undertaken in the course of preparation of the 2nd, revised edition of the academic 'Dictionary of Foreign Words'. - 1. Several entries were omitted, since the respective words (of Common Slavonic origin) had been erroneously explained as loanwords in the 1st edition. - 2. Words presented originally as polysemic had in several cases to be divided into separate entries for the reasons of etymological homonymy. - 3. In some cases, etymological information provided in the 1st edition was obsolete and/or obviously erroneous (or missing); it was made consistent with the present state of knowledge.- 4. Special attention was paid to words derived from proper names; more relevant extralinguistic information (from the field of history, geography, culture, etc.) was added in the 2nd edition.
EN
(This abstract also covers Part I of the paper published Ibid. vol. 100 (2004), No. 3. pp. 260-273). The paper discusses the etymology of the following Hungarian tree names: bükk (beech), tölgy (oak), koris (ash-tree), gyümölcsény (hawthorn, a plant similar to whitethorn or elder), gyertyán (hornbeam), gyurufa (a kind of tree similar to maple or cornel), éger (alder), dió (nut), mogyoró (hazel). The author claims that tölgy is of Iranian, more precisely Alanian-Ossetian origin, mogyoró is either of Turkic origin or an internal Hungarian development from Old Hungarian mony 'egg', while the origin of éger is very uncertain. The remaining six and a seventh item, som (cornel), not dealt with in this paper, are of Turkic origin. The first part of the paper focuses on the name bükk (beech). The author discusses its possible Indo-European origin. Further, he summarises the geobotanic background of the borrowing of tree names. He tries to reconcile the available geobotanic data with the linguistic and historical ones. In search of the Turkic background of the tree names at hand, the author also deals with some other Hungarian, Slavic, Ossetian and Turkic tree names. Finally, he points to the fact that the geobotanic map of these trees outlines the region where Hungarians may have borrowed the names, and thus this research opens a new type of source for the reconstruction of the early history of Hungarians in the first millennium.
Slavica Slovaca
|
2011
|
vol. 46
|
issue 2
156 - 160
EN
The article discusses the Balto-Slavic word *šama- ‘sheat-fish, Silurus glanis’ > Russian, Ukrainian, Bulgarian сом, Serbo-Croatian сȍ м, Slovenian sòm, Czech, Slovakian sumec, Polish sum, Low Sorbian som, Lithuanian šãmas, Latvian sams, which is problematic from the point of view of etymology. Giving an in depth analysis of previous hypotheses, the article suggests turning to the already known lexical material recorded in dictionaries. On semantic concretisation of the examples and drawing from the dialects of the Slavic and Baltic languages, the old etymology of this word (Proto-Indo- European *k’em- ‘stick, pole, horn’) becomes linguistically well-reasoned.
Slavica Slovaca
|
2011
|
vol. 46
|
issue 1
36 - 39
EN
Up to now, the etymology of Common-Slavonic *sъtъ ‘honeycomb’ is not explained sufficiently and it remains uncertain. The author of the paper discusses various existing explanations of the word, and proposes a new solution according to which the word belongs to the originally onomatopoeic root *sъp- ‘to hum, to drone’. This root is only a variant of a larger onomatopoeic set of forms that served to express different secondary meanings of the original meaning ‘to gasp, to pant, to breathe loudly, hard’ (*sop-, *sap-, *sip-, *sěp-, *sup-).
EN
The article describes a method to analyse contemporary Slovak vocabulary with regard to the origin of the words. By using statistical data from a representative corpus of modern written language and etymological information we arrive at reasonably confident estimation of the ratio of loanwords in common Slovak vocabulary and the provenance of lexical borrowings. We demonstrate some of the findings in tables and charts, providing information that is interesting to non-linguistically oriented members of Slovak population (who are sometimes vocal in expressing their attitudes to the perceived amount of loanwords in the Slovak language), but can be also inspiring for further research in philology or linguistics.
Onomastica
|
2003
|
vol. 48
159-172
EN
The article presents a semantic and structural discussion of 113 names of open fishing waters, selected from four lakes lying within the area of the district of Elk. The toponyms in this article were classified and analyzed according to the model proposed by H. Gornowicz in a work on the names of the Borzechowskie lakes in the Kociewie region. Among the names gathered and presented, the most, 93, are topographic, while the names of 14 are connected with their exploitation, and 7 are commemorative. Structural classification allows to note that compound names predominate among the names of these waters (97). Represented in the greatest numbers are prepositional phrases, with the preposition 'na', very characteristic of directional localizational toponymic names, combined with an allative accusative.
10
Content available remote

Ways of Germanisms into Lithuanian

80%
EN
Already the first Lithuanian writings and old dialects contain a number of words adopted from other languages. Researchers often disagree on the ways of their entering the Lithuanian lexicon. Some of these loanwords must have come directly from the Germanic languages, mostly German, while others were received from the Slavic languages, mainly Polish, Belarusian or some other language. In this context the problem of the intermediary language is often pertinent. In his book 'Die Germanismen des Litauischen. Teil I: Die deutschen Lehnwörter im Litauischen', K. Alminauskis recorded 2,770 words; of that number about 130 loanwords are of uncertain origin. Sometimes the ultimate form of the borrowing does not help in determining its origin, because the phonetic and morphological structure of the German and Slavic base forms is fully or nearly identical, e.g. Lithuanian 'laterna', German 'Laterne', Latin 'laterna'; Lithuanian 'gruntas', Middle Low German 'Grunt', Polish 'grunt'. These Germanisms are not old borrowings in the Slavic languages, they passed into Polish in the fifteenth and sixteenth centuries. Thus the Lithuanians and the Poles could have adopted them approximately at the same time. Some of them, lacking Polish phonetic and morphological features, could enter the Lithuanian language directly from the Germanic languages. Their sources could be various: the language and dialects of former East Prussia, Low German of Livonia, Baltic German (Baltendeutsch) language, Low German of Hansa merchants, and lastly the German language used in the Grand Duchy of Lithuania. Some words could enter Lithuanian via writings, in the translations of religious texts. It is also probable that some of them were taken over into Lithuanian directly from Latin through the works translated from Latin. They were not used in the spoken language, nor were they found in dialects. As Lithuanian written records are of comparatively late origin, the article presents some information on the history and ethnicities of Prussia. They must give a better understanding of German dialects as a source of numerous German borrowings in Lithuanian. .
EN
The paper offers detailed consideration of actional and verbal aspects of the triad of the ritualized actions «road», «crossing» and (reaching of) «lands beyond the river», which forms the basis of the East Slavic burial rite. Due to the linguistic specifics of the study the authors put their preferential attention to the language of the burial ritual, which is folklore texts (keening, spells, ritual precepts), their vestiges, separately existing as idiomatic phrases and separate words with peculiar semantics. Being the verbal counterpart of the ritualized action, textual information not only shows the interconnection between three stages of the burial ritual, but also encourages thinking about possible existence of mythological texts in the past, where the path of the dead is described – starting from gathering to the road, ritual precepts, and entering the Paradise extents.
EN
The article was inspired by two recent collective monographs (Divičanová et al., 2015, 2017) dedicated to the surnames used by the families of Slovak origin living in Békéscsaba (Békešská Čaba) and Tótkomlós (Slovenský Komlóš) as the centres of the Slovak ethnic community in south-eastern Hungary whose members started to settle in this part of the historical “Lower Land” as early as in the first quarter of the 18th century. In the above-mentioned publications, hundreds of surnames were analysed from various onomastic aspects (structural types, origin, onymic motivation, etc.). In this study, the author focuses on some questions of etymology and presents alternative explanations for about 30 surnames. In his opinion, some of the surnames under discussion may reflect autochthonous substantial formations otherwise not attested as common nouns in the Slovak language (Hamza, Holok, Lászik, Losjak, Rotyis, Szpluvák); for other surnames, the author identifies possible sources in various foreign languages historically used in this part of Central Europe, i. e., Hungarian (Balán, Csilek, etc.), Romanian (Drimba, Opra, Tirják, etc.), German (Baiczer, Gofjár, Majzik, Matajsz, etc.), Latin (Filadelfi), Polish (Kendra) and Slovene (Raspotnik).
EN
Borrowings from Indigenous American Indian languages, that got into French via Spanish is a work from historical linguistics, concentrating on the major Americanisms, originating in Indigenous American Indian languages which ultimately got into French via Spanish. The work focuses in particular on such languages as Arawak, Nahuatl, Quechua, Aymara, Araucanian, Tupi-Guarani.
EN
This paper is a sequel to an earlier one published in the same journal (1997: 207 -16), containing linguistic and cultural-historical commentaries on certain expressions in the earliest extant Hungarian-language document known as Sermo super sepulchrum (Halotti Beszéd és Könyörgés, abbreviated as HB., respectively HBK.) that has survived as copied into Pray Codex, a Latin codex from around 1195. First, it deals with the expression pur eN chomuv (uogmuc) '(we are) dust and ashes', occurring early on in HB. and referring to the transience of human life, man's insignificance as compared to God. The expression is a set phrase of Biblical origin (cf. Gen. 18. 27); it has a number of versions and can be set in parallel with other Biblical or religious expressions, too. The paper then discusses a particular three-part figura etymologica known in Hungarian mainly from HB. The Lord says to Adam (Gen. 2.17): 'for in the day that thou eatest thereof (of the tree of knowledge of good and evil) thou shalt surely die'. (The sentence is also found in HB., with the phrase corresponding to that italicised above being halalnec halaláál holA (lit. you will die a death of death)). This formula must also have become a set phrase by the time of HB., and it can also be quoted from later texts such as various points of the first Hungarian translations of the Bible, as well as from other early ecclesiastical or even secular texts. Finally, the author discusses the expression mend w NAentii el unuttei (cuAicun) '(among) all his saints and chosen ones' from HBK., corresponding in Pray Codex to inter sanctos et electos suos. The author claims that the word unuttei, otherwise unknown in Hungarian, is probably a copying error for ununei (read ününei) 'his own ones'. This is supported by theological reasoning, by citing similar Biblical or ecclesiastical expressions, and a few cases attested in early Hungarian codices where n and tt were mixed up during copying. The discussion of the three expressions shows that, by the time of HB., Hungarian ecclesiastical usage had acquired a set of phrases based on Latin texts but used (partly) independently of them, too.
Slavica Slovaca
|
2023
|
vol. 58
|
issue 1
122-137
EN
The focus of the presented paper lays on the analysis of Rudolf Skotnický's conceptual work entitled Material for an etymological dictionary, which is most likely the first comprehensive etymological dictionary in the Slovak linguistic environment. However, in his attempt to provide an etymological interpretation of individual words, R. Skotnický does not proceed in accordance with contemporary etymological knowledge, and when clarifying the genesis of words, he often starts from the sound and meaning similarity with words that have a completely different origin. This way of etymological interpretation is called folk etymology. The purpose of our work is to demonstrate folk etymological derivation using examples of specific words from the mentioned dictionary. At the same time, we compare the individual analyse with contemporary etymological works.
EN
The author discusses etymology of Latvian: 'suns' (dog; he-dog), dialectal 'suo' - as a bidding word and of archaic 'kuna' (she-dog), seeking their origin in Indo-European name of the dog: *keu-on. He stresses that according to the centum - sataem theory 'suns - kuna' represent a centum - sataem copy. It is quite obvious that there exist different semantic connections between 'centumic' and 'sataemic' words. One of the types of such connections is gender and species differentiation: 'suns' is a gender word and 'kuna' is a word describing the species, 'suns' in the meaning of he-dog describes a species of animals. The meaning of 'kuna' (she-dog) is secondary, similarly the primary meaning of '*sun(i)os' is simply dog, he-dog being secondary one. It is because of that in the Indo-European prehistoric language the names of animals mean zoological species only without differentiation of gender: males and females have no separate names. The formation of an equipollent opposition *sun(i)os (he-dog)~ kuna (she-dog) can be traced with kind-gender differentiation in the Baltic languages only, where nouns that mean creatures differentiated according to gender acquire names according to being male or female. A special diagram is presented to sum up author's views. In his opinion the existence of centum - sataemic pair in the Baltic languages has not yet proved the gender-species differentiation between centumic and sataemic word. Originally centumic and sataemic words were semantically equal. But in the course of time they develop different semantic ties due to the species-gender differentiation, which finds its expression in equipollent and in privative oppositions. The author discusses also etymology of the slavic name of dog and connects it with the Latin 'pecus' (domestic animal) and analogous forms. At the same time he does not deny the possibility that the class name dog has proceeded from the colour of the animal's fur. He also holds that the expression 'sukin syn' originally was a patronymic unit. Lexical units of the type 'Sorokin, Mukhin' are patronymic in the history of the Russian language.
Slavia Orientalis
|
2006
|
vol. 55
|
issue 3
393-399
EN
The article is devoted to the etymology and reconstruction of some Slavic dialect words (Czech 'baliga', Kashub. 'guldra', R. 'bagan', 'cheriabat', Old-Bulg. 'usma'). The analysis of these allows the reconstruction of Old Slavonic lexical units that have manifestly a dialect specifics.
EN
The purpose of this paper is to specify the chronology of the regressive rhythmic law according to which the Proto-Slovak long roots of Stang’s accentual paradigm AP (b) regularly shortened when followed by the suffix -ár. The proposed sound law operated in Proto-Slovak, before the definitive fixation of the stress on the initial syllable in Slovak and before the Slovak Rhythmic Law.
Acta onomastica
|
2009
|
vol. 50
|
issue 1
48-62
EN
In the present study the Czech toponyms consisting of the component *treb- are analyzed from the point of view of internal Czech toponymical models. In most of them the hypocoristica shortened from the binary personal names or their derivatives are identified. The remaining place names are derived from appelatives which are usually formed from the Common Slavic verb *terbiti ,to clear (a forest); to purge'. Finally the attempt to connect it with the Indo-European etymon *treb- ,to live, settle; beam; house, hut' is discussed.
Slavica Slovaca
|
2009
|
vol. 44
|
issue 1
5-12
EN
From a long time ago researches have been done in the historical linguistic and dialectology and they will always remain as a part and meaning of national identification, especially in time when the searching for national identity has got a support in the natural, uncontrolled language development and when it respects a past experience. Semantic analysis of vocabulary in dialects of particular Slavonic languages after a complex area elaboration - for instance in the Slavic Linguistic Atlas project - gives a testimony to differentiation as well as integration features of examined items with surrounding language environment. On several illustrative examples, such as 'dedina' (village), 'strom' (tree), 'hora' (forest), the author points out the constants that are permanent identifying features of Slovak language and that, in course of natural development from a supra-dialect sphere in the pre-standard period, have become constituent parts of the contemporary standard Slovak language.
first rewind previous Page / 3 next fast forward last
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.