Finnish Language

The study of Proto-Norse loanwords in the Saami languages has provided crucial insights into early Germanic-Saami contacts. However, key aspects of these borrowings, such as their relative chronology and the precise linguistic background of their source varieties, remain insufficiently explored. In this talk, I argue that Proto-Norse loanwords in Saami can be divided into at least two distinct strata, each originating from a different source language. One stratum derives from a language closely resembling Proto-Norse as reconstructed through the comparative method, while another reflects a now-extinct early Norse variety with phonological features unattested in recorded Germanic languages.

A particularly striking phonological feature in the latter stratum is the substitution of Proto-Norse postconsonantal *j with the Proto-Saami alveolopalatal affricate *ć, as seen in examples such as Saami *skālćō ‘seashell’ and *āvće̮ ‘bird-cherry’ (> North Saami skálžu, ávža) from Proto-Norse *skaljō- and *hagja- (> Old Norse skel, hegg). I propose that this pattern reflects a sound change in the source variety, where postconsonantal *j developed into a stop or affricate. In addition, this language did not exhibit Sievers’s law, which vocalized *j into *i after heavy syllables in other Germanic varieties. The presence of more than 30 such loanwords in Saami, including the Saami ethnonym for Norsemen (*tāńće̮), suggests that this now-extinct Norse variety played a major role in the earliest Norse-Saami contacts.

The distribution of these loanwords further supports a geographically localized influence. Place-name evidence from Troms county in northern Norway, such as the North Saami island name Sážžá (< *sāńćā), derived from a predecessor of the Norwegian name Senja, suggests that this extinct Norse variety was spoken along parts of the Norwegian coast. The findings presented in this paper contribute to a more refined understanding of the phonological and dialectal diversity of early Norse and its role in prehistoric language contact in Scandinavia.

In the literature and discussion about information retrieval (IR) the words "string", "word", "term" and "concept" are used inconsistently. There is no common agreement on the concepts named by these words. Thus the communication in IR is inefficient, inaccurate and prone to errors. Based on the concept of term in the study of terminology and on a three level model of IR, the paper seeks to clarify the terminology and proposes the terms string, search key and concept, with their subterms, for IR. Consequently it is proposed that a number of other words commonly used, such as "controlled term", "uncontrolled term", "free term", should not be used in IR.

Several PU words with *-mt-resemble IE ones. For *wamta 'forest', based on other ex. of *e & *i > a (https://www.academia.edu/143975134), it could be part of IE *w-w > PU *w-m or *m-w : IE *wid-won-'knowing / brain' (https://www.academia.edu/129119764) PU *wiδewe > F. yty, ydyn g. 'bone marrow / core / power', Es. üti, üdi g. 'marrow' PU *wiδeme > Erzya udem 'marrow / brain / intellect' IE *wetuso-'old' > L. vetus, OLi. vetušas; PU *wiǝtuǝšë > *wyǝtwǝšë > *watmǝšë > *wanšë 'old' IE *widhu-'wood / tree' > E. wood *widhw-aH2 (neuter plural or fem.?) > PU *wiǝtwa: > *watma > *wamta 'forest' https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/wamta IE *work^wuko-> Ar. goršuk 'badger' (https://www.academia.edu/129175453) *wërs^wuka:(y-) > PU > *märskwä > *märkä(s) \ *m\när(s)kä \ *närkäs 'badger' (below, D) The difference in *-dw-& *-dhw-might show that *dh > *th before changes to *dC. B. m, ntl,-ay Some PU words show variants with front vs. back V's. Many of them end in *-a, likely showing that PIE had fem. *-aH2(y)-from *-oiH2-(https://www.academia.edu/129368235). If so, PU *-ay > *-ä \ *-a. With some previous examples also showing IE matches, I'd add : *m(e)ntH3-uRo-> Greek μόθουρα \ móthoura 'shaft of an oar', Lithuanian mentùris 'cooking beater ', *mH3ontuRo-> Gmc *mandula-z 'handle on a grinding mill' *mentH3-tlaH2(y-) > *mentla:(y-) > *melä \ *mela 'oar' Note that there is no long **V: in Finnish (below, C). More in https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/mel%C3%A4 C. Václav Blažek in https://www.academia.edu/118037313 provides ev. for IE words with *me(:)l-'think / perceive / sense / taste'. This seems to me to show *melH1-\ *meH1l-with H-met. (https://www.academia.edu/127283240). I don't agree with his details, but he also mentioned PU *mäle 'senses / mind'. Since this appears as Finnic *meele, it would be rec. as *mäxle by those who believe in PU *x. Clearly, *meH1l-& *mäxle would be too close to simply dismiss. Again, *e > *a (fronted by *-e, maybe < *-i: or *-Vy). The opposition to *x comes from those who think many Finnic *V were lengthened before certain C(V). I find it hard to accept that *melä \ *mela 'oar' would need to behave so differently from **mäle (with similar changes in Saamic) just to get rid of *x. Most protolanguages contain much more complexity & more C's, V's, than their daughters. Why would *x, similar to (and here cognate with) IE *H be treated in exactly the opposite way? *H was doubted in the past, now is standard, but *x somehow remains controsversial? Most who oppose *x causing V: also oppose any relation to PIE, except for Hovers (https://www.academia.edu/104566591). Cognates in https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/m%C3%A4le D. w-w, rs^kw Above, I proposed : IE *work^wuko-> Ar. goršuk 'badger' (https://www.academia.edu/129175453) *wërs^wuka:(y-) > PU > *märskwä > *märkä(s) \ *m\när(s)kä \ *närkä(s) 'badger'

The paper reports on the current steps in the creation of a new corpus of written Komi-Permyak under grant number NKFIH FK 143242. The texts of the corpus are annotated by our research group using the FieldWorks Language Explorer (FLEx) software by pre-labelling sentences with the built-in general-purpose morphological parser of the software and manually checking the resulting machine-generated suggestions. As an output of the project, we will make the FLEx file available for other researchers to help them annotate their own texts. In order to use the general parser, we need to prepare it for the Komi-Permyak language system, i.e. we need to formalize Komi-Permyak morphology according to the needs of the software. After describing the basic principles of the parser, the paper presents some concrete examples of the challenges of this process in relation to adjectives and verbs.

• Can Finno-Saamic *ćć be reconstructed to account for the relationship Proto-Finnic *cc ~ Proto-Saami *ćć? • Is this supported by regular shared etymologies? • What is the background of Proto-Finnic *cc ~ Proto-Saami *ćć? • Did they emerge due to loans or can regular instances be found in inherited vocabulary as well? • Common development?

"Proto-Finnic *c in Karelian and Finnish"
Following a 1912 idea from N. Moosberg and K. B. Wiklund forgotten in later literature, I argue that North Karelian /s/ is the regular reflex of the Proto-Finnic affricate *c, retained apart from the general development *s > /š/. The contrast *c : *s can be assumed to have initially yielded a laminal : apical opposition in Old Karelian and Proto-Northern Finnic. Geminate laminal */s̻s̻/ can be moreover assumed as the initial Old Finnish reflex of geminate *cc, before its further fronting to /θθ/, judging by known dialect areas showing /ss/; old manuscript evidence transcribing ‹ss› also elsewhere; and at least one loanword into Sami showing an unexpected shibilant /šš/. Transient short laminal */s̻/ similarly allows for more natural routing of the East Finnish development *cr > /tr/ and the Southwest Finnish development ? *c / *s > /θθ/ after /i/ in a few morphological categories.

"History of Mordvin dialects in light of voiceless liquids"
Comparative-historical dialectology of Mordvinic has so far remained at middling development despite available documentation. This case study considers the rise of voiceless liquids (phonetically often characterized as only post-devoiced and here transcribed as aspirated) in the major dialect dictionary (1990–1996) of Heikki Paasonen's materials from the turn of the 20th century. This innovation, usually considered characteristic of Moksha, is present also in the Shoksha dialects of Erzya, and absent from the Tatarstan diaspora dialects of Moksha (the latter showing also other clear archaisms), indicating post-Proto-Moksha and post-Proto-Erzya spread. Southern dialects of Erzya in Penza Oblast moreover turn out to show an only superficially similar development of voiceless liquids only before dental stops t, ť, interpretable instead as evidence for the retention of earlier *RfT or *RxT clusters still until the Common Mordvinic and Proto-Erzya periods. It is hoped the results underline the importance of not neglecting dialect evidence in the reconstruction of overall Mordvinic historical phonology.

"Sorsa 'duck' and sotka 'pochard': submerged sound correspondences". The Finnish sorsa ~ suorsa 'duck', despite cognates elsewhere in Finnic, seems phonotactically non-native and can be, in the first place, analyzed as a borrowing from a Sami etymon that I reconstruct as *ćuoδšē ~ *ćuoδš-ō 'some species of duck', whence e.g. Northern duoršu, Sea Sami šuorə̆ši 'mallard', Kildin čuərššă 'teal', Southern sjotsjo 'duck'. This tentatively appears to be a derivative of a previously unidentified, possibly substratal type from a Uralic word root *ćoδ(ə) 'diving duck', well known in e.g. Finnish sotka 'pochard', formerly 'goldeneye'. Closer phonological analysis of this latter etymon also suggests the previous standard reconstruction *ćoδka must be derivative in *-ka among western languages, as indicated especially by Khanty *sāj (< unsuffixed *ćoδ´ə) and Mari *šoe (< trisyllabic *ćoδə-ka).

Résumé : On sait que le vouvoiement et le tutoiement sont des marqueurs linguistiques qui reflètent les normes de politesse et de respect dans les interactions sociales. La façon dont ils sont utilisés varie selon les langues et les cultures. L'objectif de cet article est de montrer que, loin d'être de pures « variations diaphasiques » secondaires-c'est-à-dire associées à la situation de communication-, ces marqueurs peuvent engager la dimension somatique des interlocuteurs, notamment lorsqu'ils introduisent une rupture inattendue du cadre interactionnel établi. Un tutoiement qui survient de la part d'un des interlocuteurs alors que le vouvoiement constituait jusqu'alors la norme de l'échange, et inversement, est une manière de « heurter » l'autre en impliquant le corps tout entier. Ainsi, cet événement énonciatif induira des changements proxémiques, gestuels voire esthésiques et passionnels. Il acquiert alors une fonction conative d'interpellation à l'autre. Dans ce sens, on peut considérer qu'il véhicule un « prédicat somatique » élémentaire, au sens de Jean-Claude Coquet, qui ne décrit pas le monde mais « dit le sensible » et agit sur lui, en faisant entrer la « réalité » dans le dialogue. En étudiant ces phénomènes de rupture dans les sphères francophone, italophone et hispanophone, nous cherchons à montrer que les marqueurs de politesse constituent la « peau de l'échange ».

We present a Finnish web corpus with multiple text sources and rich additional annotations. The corpus is based in large parts on a dedicated Internet crawl, supplementing data from the Common Crawl initiative and the Finnish Wikipedia. The size of the corpus is 6.2 billion tokens from 9.5 million source documents. The text is enriched with morphological analyses, word lemmas, dependency trees, named entities and text register (genre) identification. Paragraph-level scores of an n-gram language model, as well as paragraph duplication rate in each document are provided, allowing for further filtering of the dataset by the end user. Thanks to changes in the 2023 Finnish copyright legislation, the corpus is openly available for research purposes, and can also be accessed through the NoS-ketchEngine concordance tool and the dep search dependency tree query tool, all at .

Medical discharge documents are summaries written by a physician about the patient's condition and aim at transferring information to other health care personnel but also to the patient. According to the legislation, the patient should be able to understand the document. In practice, however, this has been shown to be problematic. This paper studies discharge documents from the patients' perspective and examines how they fulfil the legislation's demands on understandability. Concentrating on the vocabulary of the texts, we analyse the frequency of domainadapted terms, abbreviations and foreign words. The material consists of 23 528 heart patients' discharge documents (5 747 126 words). The analysis is performed with the morphological analyser FinTWOL (). Altogether, FinTWOL analyses 24% of the corpus as unknown or foreign words, abbreviations or medical terms. The most common category, unknown words, includes misspellings and medical terms, such as l.dex. Of these, 100 most common cover for 43% of the total. These terms thus seem to be relatively fixed. Of the words analysed as abbreviations, some are common also in standard language, but others are still very domain-specific, such as I.V. (intravenous). Also the used abbreviations are very fixed: the 100 most common ones cover for 94% of the total. This, however, does not help the patient who probably reads only one document. Similarly, even though misspellings are globally infrequent, they still occur more than once per document. In order to place the obtained results in a context, we performed a similar analysis on general Finnish university newspaper text from Turku Dependency Treebank. In comparison with the 24% obtained with the discharge documents, from the total of 10 687 words, 8,6% were given a special tag. The results show that that terms and abbreviations are considerably more used in discharge documents than in general newspaper text. It is clear that a text with such a vocabulary is domain-specific and distinct from the language that the patient is used to. Also e.g. the varying use of upper and lower case letters (dg and DG for diagnosis) emphasize the particularity of the language. In standard language texts such writing would not be acceptable. Standard writing would, however, help the patients to better understand the texts.

To the north of lake Hepojarvi there is a group of small lakes, one of which is named Laydaka [lajdaka]. The name Laydaka looks non-Finnic, especially the ending-daka [daka]. Previously it was shown that a number of hydronyms of the Leningrad Oblast, which are generally supposed to be of Finnic origin, actually have no Uralic etymologies at all but can be explained through Yeniseian roots. The component lay [laj] of Laydaka corresponds to Proto-Yeniseian * ʔm "side", the component daka corresponds to Proto-Yeniseian *deʔG "lake". The hydronym Ladoga looks much alike the hydronym Laydaka and also has no reliable etymology. The hydronym Ladoga originally also probably had the view *laʔm deʔG. Lake Ladoga in the Neolithic period was an endorheic lake; the Neva River didn't exist yet, and so the Neolithic people definitely considered the lake as a separated water basin located to the side of the Littorina Sea.

The distribution of infinitive suffix variants in North Mansi The article examines the distribution of infinitive suffix variants in North Mansi and the factors driving this morphophonological variation. The research is based on a dataset of 1,790 infinitive forms and follows a datadriven, classificatory morphophonological approach. North Mansi infinitives exhibit three suffix variants: /-ŋkwe/, /-aŋkwe/, and /-uŋkwe/. The consonantinitial /-ŋkwe/ variant occurs with irregular verbs. However, the alternation between /-aŋkwe/ and /-uŋkwe/ has not been thoroughly described before. The data reveal that this variation is a complex phenomenon influenced by two primary factors: the number of syllables in the base stem and the final derivational suffix in a derived stem. Keywords: Mansi (Vogul) language, morphophonology, infinitives, verb derivations

Vuodenvaihteessa Harvardin yliopiston presidentti erosi tehtävästään plagiointikohun saattelemana. Samaan aikaan Suomessa käytiin tietokirjojen viittauskäytännöistä vilkasta keskustelua, joka johti Journalisti-lehden päätoimittajan eroon. Näissä tapauksissa keskustelu on keskittynyt viittauksissa havaittuihin puutteisiin: kriitikoiden mukaan kaikkia lähteitä ei ollut merkitty lainkaan tai lähdekirjallisuudesta tekstiin otettujen tietojen alkuperää ei ollut tuotu riittävästi esille. Vähemmälle huomiolle ovat jääneet ongelmalliset tieteellisen kirjoittamisen käytännöt, joissa lähteet mainitaan, mutta epäselväksi silti jää, kuka on tekstissä todetun takana ja missä määrin. Pahimmillaan käsillä on unohtunutta viitettä merkittävämpi eettinen ongelma: toisten tutkijoiden suuhun asetetaan väitteitä, joita he eivät ole esittäneet. Vuodenvaihteeseen asti Media & viestintä-lehden kirjoittajia ohjeistettiin sisällyttämään kirjallisuusviite sulkeisiin virkkeiden ulkopuolelle omalla päättävällä pisteellään varustettuna, mikäli viitteen ala on useiden virkkeiden mittainen. Ohje kuului näin: Jos viittaus koskee vain edeltävää virkettä, piste ja sulut merkitään seuraavasti: (piste tulee viitteen sisältävien sulkujen perään). Esimerkki: Oppihistoria on sen tutkimista ja esittämistä, kuinka sen kohdealaksi valittu tutkimus-tai tieteenala on ajan saatossa kehittynyt (Pietilä 1997, 15). Mikäli viitataan useampaan virkkeeseen, viite merkitään viimeisen virkkeen pisteen jälkeen näin: (viite ja piste tulevat sulkujen sisään.) Esimerkki: Oppihistoria on sen tutkimista ja esittämistä, kuinka sen kohdealaksi valittu tutkimus-tai tieteenala on ajan saatossa kehittynyt. Lisäksi alkuliite oppi täsmentää asiaa: oppihistoria luotaa nimenomaan ajattelun ja tutkimuksen kehitystä. (Pietilä 1997, 15.) Sama viittauskäytäntö esitellään monissa suomalaisissa tutkimusoppaissa, kuten tutkielmien teossa paljon käytetyssä Hirsjärven, Remeksen ja Sajavaaran teoksessa Tutki

DIGISTI PAREMPI KEHITÄ DIGITOINTIA VASTUULLISESTI Julkaisu perustuu Joustavat koulutus-ja työelämäpolut tulevaisuuden digitointiosaajille (JoDi) -hankkeen Digitoinnin kehittäjä -koulutukseen, joka toteutettiin Kansallisarkiston ja Xamkin yhteistyönä 2024. JoDi-hanketta rahoitti Etelä-Savon ELY-keskus Uudistuva ja osaava Suomi 2021-2027 EU:n alueja rakennepolitiikan ohjelmasta (ESR+). Hankkeen toteuttivat Kaakkois-Suomen ammattikorkeakoulu Oy, Itä-Savon koulutus kuntayhtymä ja Kansallisarkisto 1.5.2023-31.5.2025. © Tekijät ja Kaakkois-Suomen ammattikorkeakoulu Tämä teos on kuvia lukuun ottamatta lisensoitu Creative Commons Nimeä-JaaSamoin 4.0 Kansainvälinen -lisenssillä (CC BY-SA).

Kansalliskirjaston digitointiohjelma vuosille 2025–2028 huomioi kestävän kehityksen tavoitteet. Nelivuotisilla digitointiohjelmilla Kansalliskirjasto pyrkii varmistamaan, että digitoitavan aineiston valintaa tehdään suunnitelmallisesti ja yhteiskunnan eri ryhmät huomioiden.

A.  Holopainen (2020) lists several IE loanwords in Finnic (sometimes also other Uralic) with met. to avoid *Tr (S. á-kṣetra- ‘destitute of fields’ >> *akštera > F. ahtera ‘barren/sterile [of cows]’, Mv. ekšt’eŕ; S. dhruvá- ‘firm / secure’, Av. drva- ‘healthy’ > Fc. *terves > F. terve ‘healthy / whole’).  If there was no met., other ex. have PU *kC replacing IE *TC :

Gmc *nēþla-, E. needle >> Fc. *nekla

S. dātrá- ‘share’ >> F. takra ‘piece of meat (as bait)’, Izh. tagr, Sm. *tāvrō

*bhNg^hu-s > G. pakhús ‘thick’, S. bahú-, Ir. *badzu- >> *patsu > F. paksu ‘thick / dense’, Mv. pokš ‘big / adult’


However, there are no other ex. of PU *ks replacing Ir. *dz despite many cases in which this would be expected, if real.  I do not see this as PIE *g^h > Ir. *dź > *dz at all, since the real PIE form was likely *dbhmg^hu-s, with met. in *dbazu- > NP dabz (Pronk, Whalen 2024b).  This makes it likely that a language with a different type of met. had *dbazu- > *badzu-, with a real C-cluster that could show *TC > *KC, not an affricate :

*dbhmg^hu-s > G. pakhús ‘thick’, S. bahú-, *dbazu- > NP dabz, *bhaγu > Kv. bok ‘enough’, *bhaRu ‘much / many’ > Bn. bɔr-, Ks. bo, *bǒṛù > Bu. buṭ (loan), *bṛǒù > Bs. ḍẓóo, ?. *dbazu- > *badzu- >> *patsu > F. paksu ‘thick / dense’, Mv. pokš ‘big / adult’
*dbhng^hulo- > G. pakhulós, S. bahulá- ‘thick / spacious/abundant/large’, A. bhakúlo  ‘fat/thick’, Ni. bukuṭa ‘thick [of flat things]’, Rom. buxlo ‘wide’
*dbhmg^hos- > Av. dǝbązah- ‘height / depth / thickness?’


With this, I see PU *waksa ‘span’ as related to Wan. lwast, etc.  Since S. vítasti- is masc. but several of these later words are fem., an ending *-a: >> PU *-a makes sense.  It could be *wlasta: >> *watsa > *waksa.  I can’t know which IIr. language was the source, but the ones beginning with wl- or *wl- > l(h)- would be most likely :

*wi-tns-ti- > S. vítasti- m. ‘span’, Snj. wiast, Lauṛ. wəryahás \ wrias, Weg. wərəst, Av. vītasti-, P. badast, Dari belest, Os. wydisn(y) \ udzesnä, *wilašt > Ps. wlešt, Wan. lwast f., *γWlašt > A. lhaásṭ ‘plain’, Kh. làšṭ ‘flat / plain/wide flat place’
?; PU *waksa ‘span’, Mv. vaks, F. vaaksa, Es. vaks \ vass \ vaaks, Sm. *vuopsē > NSm. vuoksi, l. vuoppse, vuok'se -vs- ‘distance between the extended thumb and the extended forefinger or middle finger’

PU *waksa but *wapsa > Sm. *vuopsē suggests that *ts > *ps near labials was also possible, maybe in PU, unless also for later *ks > *ps near labials.  This seems to match n > m by P, also in loans (2025a) :

*mliHno- > W. blin ‘tired’, Lt. blīnis ‘tired person’, OCS mlinŭ, R. blin ‘(thin) pancake’ >> Ud. mil’ym \ bil’ym


B.  All these changes have implications within PU.  Knowing that some *ks came from *Ts makes looking for cognates in native words easier.  One good ex. is PIE *H3ozdo-s ‘branch’ > PU *oksë ‘branch’, which includes *H3 > *w, *H1 > *y (Whalen 2025b) for *H3o- > *wë- > *(w)o- :

*H3ozdo- ‘branch’ > G. óz[d]os, Aeo. úsdos, Ar. ost, Go. asts, OI odb ‘knot/outgrowth’, W. oddf, *oz(ä)do- > *esäle > TA asäl, TB esale ‘post’
Mac. ázox [*-s] ‘wood’, Mac. áddai p. ‘logs for fuel’
*wozdo > *wëdzë > *wotsë > PU *oksë ‘branch’ > F. oksa, Z. vos, Ud. usy ‘harrow’, Mr. ukš


With *e > *iǝ (like Tocharian), then stressed *iǝ > *ǝ > *a but unstressed > *i (Whalen 2025b), I assume that prepositions weren’t stressed to explain :

*trH2ants > L. trāns
*terH2ants > *tirxats > *tariks ‘across’ > Mh. turks, Mv. troks, Mr. toreš


This might exist (with several alternatives) in :

*k^oH3t- > L. cōt- ‘whetstone’, *k^awt- > cautēs ‘rough pointed rock’, *k^H3to- > catus ‘sharp/shrill/clever’
*k^oH3t-isk^e- ? > *koitske- > PU *kekskV ‘sharpen / whet’ > *keskV \ *keksV, Z. keslï-, Mi. kiwt-

but it could be directly from the root in *k^oH3-isk^e- > *koixske- > *koikske- > PU *kekskV.


C.  Another odd shift might be seen in (Helimski et al.) :

Number: 189
Proto: *jokse-
English meaning: to run; be rutty, copulate
German meaning: laufen; ? läufig od. brünstig sein, sich paaren
Finnish: juokse- 'laufen; fließen; brünstig sein' ?
Estonian: jookse- 'laufen, rennen; belaufen; (von Fischen) zum Laichen dem Flus hinaufgehen' ?
Saam (Lapp): juk'sâ- ~ jǫk'sâ- -vs- (N) 'reach, overtake, obtain', jåkså- (L) 'einholen, gelangen bis', juksi̊- (T), juχse- (Kld.) 'einholen, erreichen'; jǫχs̨e- (Ko.) 'verschwinden, sich entfernen' ?
Hungarian: ív-, dial. iv-, ví-, vív-, vij- 'laichen (Fisch)'


Even with a fair amount of uncertainty, I would say that the 2 meanings seem related to Ir. *yamP- ‘move / wander / etc.’ (Cheung) vs. S. yabh- ‘fuck’ (whose relation to each other is disputed, but if PU *jokse- can mean both, why not?) :

*yeP- ‘arrive / move / depart’ > TA yäw-, TB yäp- ‘enter / set [of sun]’, Lw. *iba- ‘west’
*yoPmo- > T. *yepme > TA yokäm ‘door’, TB yenme ‘gate/entry/portal’
*yeP-ne- > *yamP- > MP jumb- ‘move’, NP junb- ‘move/stir/shake (intr)’, Sg. y’β- ‘wander/travel/rove’, Mj. yōb- ‘dance’
*ymp-sk^e- > S. yucchati ‘go away / depart / keep aloof / vanish  / *stray from the path > err’

*H3yebh- \ *H3oybh- ? ‘fuck’ > S. yabh-, G. oíphō
*H3yebh-sk^e- > Ir. *yaPsa- >> Ar. yawsem \ yus(n)um ‘fuck / deflower’
S. *yápśma- > yákṣma- ‘~ disease’ ?? (if a sexual disease or impotence ?)

If related, *e > *iǝ, stressed *iǝ > *ǝ > *o by P (before *psk > ks, or similar).  All could come from *H3yebh- just as easily as from separate *(H)yeP-.


D.  Tocharian *-ts > *-ks ( > -k ) & *-ts- > -ps- in (Whalen 2025c) :

*H1etsonyo- > *H1yetsono- > *yets(on)o+nt- > *yätsent- > TA yäpsant ‘autumn’

*paH2ant-s > G. pâs, pan(to)-, ‘all’, *pānts > *pānks > T. *pōnxs > TA puk, pont p., TB po, ponta p.

also match these PU changes.  I have said that Tocharian was close to PU (Whalen 2024c), and add these ideas in support.  Loanwords here & there are unlikely due to timing and the other changes seen within PU.  Obviously, trying to find an Indo-Uralic level for all these would be hard.  No information within PU does not exist in PIE, so an IE > PU path is needed.  There is no *-CC that is likely to become *-ks in PU, but *-ts in PIE, with later *-ts > *-ks in a sub-group like Tocharian.  Even using derivatives of *terH2- ‘cross’ for ‘across’ would not last over 10,000 years in a supposed PIU, if *tarits > *tariks ‘across’.

I think that knowing the PIE sources can explain some irregularity in PU.  It also allows an explanation for various Uralic features of a disputed nature.  Whether it had *-x- after many V’s to create length in Finnic and *Vǝ in Samoyed can be seen by observing the IE sources :

*sk^(e)HyaH2 ‘covering / shadow’ > TB skiyo, G. skiā́, NP sâya ‘shadow / shelter’, *sk’iǝx’ya: > *sx’iǝx’ya: > PU *saxja ‘shadow’ > F. suoja ‘shelter / refuge / protection , Ud. saj ‘shadow’ , Z. saj ‘shelter’

*polH2o- > OCS polŭ ‘side/shore’, *pelH2o-m > *piǝlxon > *palxöy > PU *päxle > Mh. päl’ ‘side’, Sm. bælle ‘side/half (lengthwise)’, F. -pieli

*polH2aH2 > SC póla ‘half’, PU *pexla:y > *päxlä > Mh. päl’ä ‘half’, Hn. fél, F. puole-

*gWelH- > OE cwelan ‘die’, *kwiǝxl- > PU *kaxle- > F. kuole-, Hn. hal


Some say that *saxja is a loan from Ir., but even if it were, the fact that *H lasted so long in Ir. (Kümmel) would still explain V: vs. V in the same way.  Native or not, it supports *-ax- or other *-aC- as the source of this alternation.  If native, it would also show asm. of *k’-x’ > *x’-x’ (or similar).  Hovers ideas make it likely that normally *sk’ and *sk (before front) > PU *c’ instead.  Since *-s- > *-x- in most environment, PIE *s can have the same effect.  Other *VCC could also act like *VxC, becoming Fc. *V:C, like *-Ctl-, *-Cl-, etc. :

*nod-tli- > Lt. našl̨i ‘reed’, *noCl’iǝ > PU *n’ëCle > F. nuoli, nuole-, Mv. nal, EMr. nölö, SMi. ńėl, Hn. nyíl, nyilat a. ‘arrow’
(meaning:  *nodo- > H. nāda\i- ‘reed / drinking straw’, *nedo- > Ar. net -i- ‘arrow’, Pth. nad ‘pipe, flute / cane, rod’)

*p(a)H2k^tlo- > L. pālus ‘stake’, paxillo- dim., *pax’tlo-m > *paytloy > PU *pexle > F. piele- ‘(door)post/doorjamb / mast’, Mh. päl’, Hn. fél-fa

*mntis > S. matí- ‘thought/intelligence/worship/desire’, L. menti-, E. mind, Li. mintìs ‘thought/idea/meaning’
*mǝntiǝ > *mantǝy > PU *mänle > Fc. *meeli > F. mieli ‘reason/understanding’

*k^romusyo- > *c’lomwǝxyo > *c’δomǝxoy > *δyëmxey > PU *δ’ëxme ‘bird cherry’, F. *toome- > tuomi (Whalen 2025a)


Hovers compared PU *mänle (his *mäli ‘mind’) to PIE *mel-, Lw. māl ‘thought / idea’, G. mélō ‘care for’.  However, the existence of *HC & *CC in all other ex. makes *-nt- > *-nl- a better source.  For PU *T > *l, see (Whalen 2024a).


As more evidence, *H(y) & *K can merge as *xj (or maybe *x’) between V’s.  Whatever the exact nature, it also seems to front & raise V’s (*e > *i, *o > *e) :

*wog^h(eye)- ‘lead / draw / pull’ > PU *wexje ‘to take/grasp’ (Hovers:  *weg^h- > *wejxi)

*weg^h- ‘lead / draw / pull’ > PU *wixje ‘to bring / take swh.’ (H:  *weg^h- > *wijxi)

*kseH(y)- ‘heat / burn’ > *kxexy > PU *kixje ‘heat / to be in heat’, Fc. *kiimä ‘heat / rut’, Ud. kemdź ‘ignite’ (H:  *gheyg^h- ‘lust for’ > *kijxi)
(S. kṣā́yati ‘burn’, kṣā́tí- f. ‘singeing / heat’)

Hovers separated his *wx & *jx from *x by their effects.  Here, his *jx > Sm. *k, Pm. *j, Fc. long *V.  Since PIE had ablaut in *wog^h(eye)- vs. *weg^h-, the PU words also identical but for V almost must be related.

Helsingin yliopiston suomalais-ugrilaisen kielitieteen henkilökohtainen ylimääräinen professori Artturi Kannisto (12.5.1874Kannisto (12.5. -10.3.1943) muistetaan parhaiten mansin tutkijana, joka vietti viisi vuotta nuoruudestaan läntisessä Siperiassa keräten kaikista mansin murteista laajemmat kieliaineistot kuin kukaan ennen häntä tai hänen jälkeensä. Suomalais-Ugrilaisen Seuran lähettämistä tutkimusmatkailijoista yksikään ei ylipäänsä kerännyt yksittäisistä kielistä tai kielihaaroista niin laajoja kokoelmia kuin Kannisto mansista (Salminen 2008: 75-76). Suuren osan loppuelämästään Kannisto työskenteli keräämäänsä materiaalia tutkien ja muiden saataville toimittaen, mutta hän toimi myös suomen kielen tutkijana, opettajana, huoltajana ja kehittäjänä. Kaikki aiemmat Kanniston elämäntyöhön keskittyneet esitykset ovat silti tarkastelleet ennen kaikkea hänen merkitystään mansin tutkijana. Kanniston fennistisiin saavutuksiinkin on kiinnitetty huomiota, mutta paljon hajanaisemmin, joko sivurooleina kuuluisan mansin tutkijan elämässä (esim. Ravila 1943; Liimola & Kulmala 1974: 23-24, 33-35) tai pieninä osasina fennistiikan, etenkin Kotikielen Seuran ja suomalaisen kielenhuollon historiaa (Paunonen 1976; Juusela 2006; Kolehmainen 2014). Tässä kirjoituksessa keskityn ennen kaikkea Kanniston näkyvimpään fennistiseen perintöön eli hänen omiin julkaisuihinsa suomen kielen tutkimuksen alalla. Koska Kanniston elämäntyössä on erotettavissa selvästi erilaisia aikakausia, kirjoitukseni etenee lähinnä kronologis-temaattisesti. Lyhyen yleiskatsauksen jälkeen tarkastelen ensin nuorta Kannistoa kotimurteensa tutkijana ja Siperiasta palattuaan enemmänkin ahkerana kielenhuoltajana. Erillisen huomion ansaitsee

Uralic names for trees can exemplify many sound changes.


A. *o > *ë

PU *ë or *ï is a phoneme sometimes acting like *ë, sometimes *ï.  In many branches *ë merged with *a.  This clearly is paralleled by Zhivlov’s -a1 & -a2 (Whalen 2025a).  Since these also merged but had differing effects in *V-a1 vs. *V-a2, it makes sense that *a1 = *a (became Smd. *-å, caused X. low V) and *a2 = *ë \ *ï (as *-ë it became Smd. *-ǝ, as *-ï it caused X. high V).  I think PU *ë is earlier, if from PIE *o in most environments.  It is the difference between a1 & a2 that corresponds to PIE *-a: & *-os.  Hovers has given many ex. of PIE *o > PU *ë (or *ï) in (with my modifications & added ideas) :

*(s)t(o)rgo-s > G. tórgos ‘vulture’, Gmc *sturkaz > E. stork, ON storkr
*torgaH2- > H. tarlā ‘stork’, PU *tërka ‘crane’ > Z., Ud. turi, Hn. daru, Mi. *tï:rïɣ > Mi.s. tāriɣ, X. *tārəɣ > .v. tarəɣ

*krokiyo- > Ct. *korkiyo-s > W. crechydd \ crychydd ‘heron’, Co. kerghydh
*korkoy- > PU *kërke (below)

*lendh- ‘to lower oneself’ > Li. lį̃sti, lendù ‘crawl / creep’
*londho-m  ‘lowland’ > Gmc *landaN > Go., E. land
*londhon > *londhoy > PU *lënte  ‘lowland’, Fc. *lanci ‘lowland’, Mr. landaka ‘small valley’, Z., Ud. lud ‘field, meadow’, Smd. *lïntə̑ ‘plain, valley’

*luH1mn > G. lûma ‘dirt / filth’, Al. (l)lym ‘silt / mud’
*lowHmo- > *lowHwo- > *loHwo- > PU *lëxwë > Fc. *liiva ‘mud, sludge, slime, sand, gravel’, Z., Ud. luo ‘sand’, X. *lïwï > .k. ḷŏwĭ ‘sludge, mud’

*loH3w- \ *lowH3- ‘wash’, Ar. loganam ‘to wash/bathe’, L. lavāre ‘wash / bathe / moisten’
*lëxwV > PU *lëkaw ‘wash’ > Fc. *liko- ‘to soak, to get wet’, Mi. *låwt- > .s. lowt ‘to wash’, X. *loɣī̮t > .v. lŏɣi̮t


To these I’d add :

*k^osnaH2- > Sl. *sosna ‘pine’

*k^osko-s ‘pine cone’ > Sp. cuesco ‘stone of a fruit’. G. kókkos ‘kernel/grain/seed / kermes oak’, kókkalos ‘kernel of a pine cone’
PU *sïksï \ *sëksë ‘cedar / (Siberian) pine’ > Ud. susï-pu ‘juniper’, Z. sus(k)-, X.v. li̮ɣǝl, Mi.km. tē̮t, Nenets tideʔ, En. tydiʔ, Skp.n. ti̮ti̮k, Kamass tēdǝŋ, Mat. tidamь, Tay. tideŋ


It is also likely that *omC > *umC, similar to opt. *orC > *urC in :

*krokiyo- > Ct. *korkiyo-s > W. crechydd \ crychydd ‘heron’, Co. kerghydh
*korkoy- > PU *kërke > Sm. *kuorkë > NSm. guorga, Mr.m. karga, karkt p., Mv. kargo, -t p.
*korkoy- > PU *kurke > F. kurke- ‘crane’, Smd. *kǝrö(-kǝrö) > Nga. kokərɨ, En.f. kori, Nen.f. kaqłyu, .t. xăryo, Skp. *qara > .n. qara, .s.N. kará, .s.U. kaara, Kam. kʰuruʔjo, Koib. kurerok, Mator körüh \ köröh


Also, this allows the many PIE *-oC to become PU *-oy > *-öy > *-ey > *-e.  This path allows changes to *-V-e to make more sense.  Zhivlov’s statement that :
>
…in Saami and Mordvin… The highly idiosyncratic nature of these sound laws, especially of the development *a-i >*o-a, makes it unlikely that the set of changes listed above occurred independently in two different languages.
>

This is not just "highly idiosyncratic", it's nearly impossible.  His *-i, as in *weti 'water' would be my *wete < PIE *wodo:r.  It would come from PIE *-o:r > *-ö:y > *-e, with a stage like *o-oy > *o-öy > *e-ey in my theory.  With one type of V-asm. from *-e, it makes much more sense that standard *a-e was really *a-öy > *ɔ-öy > *ɔ-ɔy > *ɔ-ay > *ɔ-a > *o-a (or similar) in Saami and Mordvin.


B.  *-oC > *-oy

I’ve said that many IE yo-stems became *-oy in PU (similar to Ar. *-yo- > *-oy- > -u- when unstressed), changing > *-e as above.  Other PIE *-oC became PU *-oy, like :

*wodo:r > *wodo:y > *wödöy > PU *wete ‘water’


Combining these, several other *-Co- > *-oC > *-oy :

*bhowHmo- > Gmc *bauHma- ‘tree’
*bhowHom- > *puwxon > *puxoy > *puxe > F. puu ‘tree / wood’, Hn. fa

*k^oH3no-s > G. kônos ‘(pine-)cone / spinning top? / bullroarer?’, S. śāna-s / śāṇa-s ‘whetstone’
*k^oH3no- > *k^oH3on- > *kuwoy- > PU *kiwe ‘stone’ > F. kivi

The reasoning for ‘stone’ (like Hovers, with a different PIE original), also seen in *k^oH3t- > L. cōt- ‘whetstone’, *k^awt- > cautēs ‘rough pointed rock’, see D.  For *H3 > *w, see D.  A similar change, with the original form slightly unknown in :

*sH2ay-mn > Greek haîma ‘blood’, *sH2ay-nes-? > Latin saniēs ‘ichor / pus’, *sH2ay-no-? > *säyon > PU *säje ‘pus’


C.  compounds

In compounds of clear origin, the needed sound changes can be examined and later applied to other cases.  PU supposedly had 2 groups for ‘alder’, but their great similarity makes that nearly impossible.  The difference seems to be that one had an early compound with *puxe ‘tree’ that underwent sound changes, the other a late (& optional) compound with *puxe ‘tree’ that did not :

PU *läl(-puxe) > Pm.*lɔ̇l, *lȯlpu > Ud. lulpu, Z. lolpu >> Mr.bk. lül-pe ‘alder’

*läl-puxe > *lälpxe > *leppä > F. leppä ‘alder’, Mv. l’epe, Mh. l’epä

I think it’s likely that *-px- > *-pp-, but dsm. of *l-l could leave a mora filled *lp > *_p > *pp instead.  When both words contain *lV()p()V, and the V’s could also match if due to met., it would be foolish to separate them without examining how many later Uralic ‘_-tree’ are already known to have *-puxe.  I’ve said that other *-V- > -0- in cp. (2025b) :
>
C.  PU *wixte is used for both ‘5’ & (in Smd.) ‘10’.  I think this is similar to PIE *penkWe ‘5’, which ends in *-e (which would be the dual ending if from a stem *penkW-, with no other reasonable source in nouns).  I’d expect a dual to be ‘both hands’ in this situation (Whalen 2025c).  If its meaning ‘all’ could apply to either ‘all (5) of one hand or / both hands (10)’, it would match Uralic *wixte ‘5 / 10’.  At an early stage, the largest number with a “simple” name being the end of a 5 count or 10 count seems to fit.  With this, an origin in *dwi-käte ‘2 hands’ (*käte > F. käsi ‘hand / arm’) makes sense.  However, instead of standard *käte, *xäte would fit better to get *-x(V)t-.  For PU *x > *k as optional, see also :
>


The 2 l’s here match those in cognates of *H1olsno- > L. alnus, Li. ẽlksnis \ ãlksnis ‘alder’ (2025d, e) :

*H1ol-H1l-mo- > *olmos > L. ulmus ‘elm’, Gmc *al(il)ma- > ON álmr, L. >> NHG Ulme
Gmc *alilmo:n- > *a_ilmo:n- > *amilo:n- > ON Em(b)la
*H1el-H1l-mo- > Sl. *(j)ĭlĭmŭ > R. ílem, íl’ma g. ‘mtn. elm’, Ct. *elilmo- > Gl. Lemo+ \ Limo+, MI lem, I. leamh, *leimo- > W. llwyf p., Gmc *ili(l)ma- > E. elm, OHG elm-boum, MHG ilm, ?Lus. >> Sp. álamo ‘poplar’

*H1le-H1l- > H. alil- \ alel- ‘flower / bloom’, alaleššar ‘meadow’, *ley-lo- G. leírion ‘lily / narcissus’, L. līlium, etc.


With this, I think it is very likely that a change lik *aliǝl > *älil > *läli > PU *läl(-puxe) took place.  The change of *e > *iǝ (like Tocharian) is needed when stressed *iǝ > *ǝ > *a but unstressed > *i.


D.  *H3 > *w, *H1 > *y

In the changes for :

*k^oH3no-s > G. kônos ‘(pine-)cone / spinning top? / bullroarer?’, S. śāna-s / śāṇa-s ‘whetstone’
*k^oH3no- > *k^oH3on- > *kuwoy- > PU *kiwe ‘stone’ > F. kivi

it seems that *H3 > *w, and *uwV remained when *uwC > *uC.  Later, some *u > *i before labials, like *lupša vs. *lipsa (2025f) :

PU *lupša ‘dew’ > EMr. lups, WMr. lypš, Mv. läkš \ lekš ‘hoar frost’, Mh. leš, Smd. *jəptå ‘dew’ > Nga. djebtua, En.f. djota, Nen.f. dyăpta, t. yăbta, Skp. *ťaptə > n. ćapty, s.N. čapt, s.U. tjapt, Kam. ʒ́eʔbda, Koib. ǯibda, Mat. čibtal
PU *lipsa ‘dew’ > Sm. *lëpsē > i. lapse, NSm. laksi, SSm. lepsie
PU *lüpsä > F. lypsä-ä, Estonian lüps-ma, Sm.t. lapse- ‘to milk tr. / yield milk intr.’, NSm. lak'câ, lāvcâ(C)- ‘cream’, *lovsə n. > Mv. lovso, Mh. lofca ‘milk’


I also see parallel *H1 > *y in :

*pelH1waH2- > Os. farwe \ färw(e) ‘alder’, OHG fel(a)wa ‘willow’, NHG Felber
*palywa > PU *playVw > F. paju, *bad’ > Ud., Z. bad’ ‘willow’, Hn fagyal, -ok p. ‘privet’, Nen. p’ew ‘inner willow bark’, Skp.s. pêê ‘bark’, Kam. po ‘linden bark / willow branch’


These are related, as other ‘pale’ colors to ‘willow’ in other IE, to :

*pelH1- / *palH1- ‘grey < dust / ash / meal’

*pelH1tno- > palitá- ‘aged/old/grey’, G. pelitnós [also lH1 > ly > li or similar]


The met. *palywa > *playVw (or later > *payVl in Proto-Hn., Hn. -l ) might explain *pl- > *bl- > *b- in Pm.  There are likely several causes, and it seems (from Hovers’ ex., if all true) to be common for *C- to voice when *H, *N, or *r occurred later in the word.  Maybe also with his *H- > *k-, instead > *g- in Pm.  I hope to examine all cases later, & see if their ety. are all true, since counterexamples might exist for some if not.


*H3 > w is also seen in many other words in IE (Whalen 2025g, Note 1), including :

*k^oH3t- > L. cōt- ‘whetstone’, *k^awt- > cautēs ‘rough pointed rock’, *k^H3to- > catus ‘sharp/shrill/clever’

*plew- \ *ploH3- ‘flow’, Gmc. *flōanaN ‘flow’, Go. flōdus m. ‘river’, E. flood

*troH3- > G. trṓō \ titrṓskō ‘wound / kill’, *troH3mn \ *trawmn > trôma \ traûma ‘wound / damage’

*sk^oH3to- / *sk^otH3o- / *sk^ot(h)wo- > OI scáth, G. skótos, Gmc. *skadwá- > E. shadow

*lowbho- ‘bark’ > Al. labë, R. lub; *loH3bho- > *lo:bho- > Li. luõbas

*doH3- \ *dow- ‘give’
*dow-y(eH1) >> OL. duim sj., G. duwánoi op. (with rounding or dialect o / u by P / W, G. stóma, Aeo. stuma)
*dow-enH2ai > G. Cyp. inf. dowenai, S. dāváne (with *o > ā in open syllable), maybe Li. dav-
*dow-ondo- > CI dundom, gerund of ‘to give’
*dH3-s- ao. > *dRWǝs- > *dwäs- > TB wäs-


Other ex. of *H1 / y :

*H1ek^wos > Ir. *(y)aśva-, L. equus
*yikwos > *hikpos > LB i-qo, G. híppos, Ion. íkkos ‘horse’
Ir. *(y\h)aćva- > Av. aspa-, Y. yāsp, Wx. yaš, North Kd. hesp >> Ar. hasb ‘cavalry’

*H1n- > *yn- > *ny- > ñ- in *Hnomn ‘name’ > TA ñom, TB ñem, but there are alternatives

*sH1emH2- > Li. sémti ‘scoop / pump’, *syemH2- > *syapH2- > Kh. šep- ‘scoop up’

*suH1- ‘beget / give birth’ >>
*suH1ur-s > *suyu-s > G. Att. huius, [u-u > u-o] huiós, [u-u > o-u or wä-wä > o-u] *soyu > *seywä > TA se , TB soy, dim. saiwiśk-
*suH1un- > *seywän-ikiko- > TB dim. soṃśke
*suH1un- > *suH1nu- > S. sūnú-, Li. sūnùs
*suH1nu- > *sunH1u- > Gmc. *sunu-z > E. son

*dhuwH1- ‘smoke’ > G. thúō ‘offer by burning / sacrifice’, thuá(z)ō ‘smoke / storm along / roar/rave’, LB *Thuwi:no:n \ tu-w...

A.  I’ve said that Uralic had *ks > *kš and other words make it clear that *r also sometimes caused ret., even at a distance, just as in Indic (Whalen 2025b) :

*ser- ‘flow’, *seraH2- > PU *sara \ *šara ‘flood’ > Mi. *tūr, X. *Lār, Hn. ár

However, there is another set of words that also show PU *s vs. *š :

PU *lupša ‘dew’ > EMr. lups, WMr. lypš, Mv. läkš \ lekš ‘hoar frost’, Mh. leš, Smd. *jəptå ‘dew’ > Nga. djebtua, En.f. djota, Nen.f. dyăpta, t. yăbta, Skp. *ťaptə > n. ćapty, s.N. čapt, s.U. tjapt, Kam. ʒ́eʔbda, Koib. ǯibda, Mat. čibtal
PU *lipsa ‘dew’ > Sm. *lëpsē > i. lapse, NSm. laksi, SSm. lepsie
PU *lüpsä > F. lypsä-ä, Estonian lüps-ma, Sm.t. lapse- ‘to milk tr. / yield milk intr.’, NSm. lak'câ, lāvcâ(C)- ‘cream’, *lovsə n. > Mv. lovso, Mh. lofca ‘milk’

I include all these groups both because some IE show the same shift (OI bann(a)e ‘drop’, bainne ‘milk’) and because *ps (& especially *pš) are rare in PU, making it highly unlikely that words for ‘drop’ & ‘milk’ would contain them & be nearly identical.  A change with dsm. of *-upC- > *-ipC- seems likely in Sm.  For IE fem. in *-aH2(y)- > PU *-a \ *-ä, see (Whalen 2025a), with other ex. in later drafts.


If the *s vs. *š in *sara \ *šara is related, it would require either *lrupsa or  *lurpsa.  These may seem like impossible forms, but with *T > *l (Whalen 2024a) it is possible *lrupsa < *dhrupsa (*d(h)r- > lr- also in Bc.) or similar.  Of course, all these words with -ps- or *-pš- resemble (much more if *dhr-) IE ones :

*dhro(w)bso- > S. drapsá- ‘drop of liquid’, G. drósos ‘dew’

*dhrewb- > ON drjúpa, dropi, OE dryppan, dropa, E. drip, drop, G. thrúptō ‘break into pieces’

The loss of *w in *-wP- is seen in many other IE words (1), G. *ps > (s)s in some (2).  I wouldn’t think that IE & PU would both have *-ps- in ‘a drop’ by chance, when relatively rare in both.  By combining these ideas :

*dhro(w)bso- > S. drapsá- ‘drop of liquid’, G. drósos ‘dew’
*dhrowbsaH2(y)- > *lṛupsa: >  *lṛupṣa: > PU *lupša ‘dew’ > EMr. lups, WMr. lypš, Mv. läkš \ lekš ‘hoar frost’, Mh. leš, Smd. *jəptå ‘dew’ > Nga. djebtua, En.f. djota, Nen.f. dyăpta, t. yăbta, Skp. *ťaptə > n. ćapty, s.N. čapt, s.U. tjapt, Kam. ʒ́eʔbda, Koib. ǯibda, Mat. čibtal
PU *lipsa ‘dew’ > Sm. *lëpsē > i. lapse, NSm. laksi, SSm. lepsie
*lṛupsa:y > PU *lüpsä > F. lypsä-ä, Estonian lüps-ma, Sm.t. lapse- ‘to milk tr. / yield milk intr.’, NSm. lak'câ, lāvcâ(C)- ‘cream’, *lovsə n. > Mv. lovso, Mh. lofca ‘milk’


B.  There is no internal problem with :

PU *nuks’e > Fc. *nuksi > Es. nugis, NX. njuhës ‘sable’, Hn. nyuszt ‘pine marten’, nyest ‘beech marten’, Ud., Z. ńiź, Nen. *nokå > noxo

but an external problem, if related to PIE, is what words it could be cognate with.  The group of  S. nakulá- ‘mongoose’, Ir. *nakuðá- > Xw. nkδyk ‘weasel’ (with some *l > Ir. ð, *kul-ōwyo- > *kulāw(w)a- ‘nest’ > Kd. kulāw, *kulāma- > Bal. kuδām, NP kunām) resembles it most closely, but its origin is disputed.  If these 2 groups look similar yet are isolated, a theory of PU >> IIr. or some substrate, etc., would work just as well, but give no information about old levels in either.  I’ve said  (2025d) that it doesn’t look like any IE root because of met. from *leH1k- \ *lek(H1)- (Nw. lakka ‘to hop / patter about’, MHG lecken ‘hop’, Lt. lḕkt ‘to spring/jump’), with a shift :

*lekHuno- ‘nimble animal’ > *nekHulo- > S. nakulá- ‘mongoose’, Ir. *nakuðá- > Xw. nkδyk ‘weasel’

PU might support this if it also had met., but after palatalization of *l, if *kl’ > *ks’.  Since no clusters like *kl’, etc., seem to have existed in PU, they would need to have undergone some sound change (if I’m right in PIE > PU).  Maybe :

*lekHuno- > *liǝkxwǝno > *l’ǝkxwǝno > *nǝwǝkl’ox > *nǝwǝks’ox > *nǝwǝks’oy > PU *nuks’e


Since the type of metathesis could be almost anything, and few ex. exist of most PU sound changes, maybe instead the palatalization moved instead :

*lekHuno- > *liǝkxwǝno > *l’ǝkxwǝno > *nǝwǝkx’ol > *nǝwǝks’oy > PU *nuks’e

It would be hard to be sure without having a better understanding of PIE > PU, and more ex. of each type of sound change.

In the fi eld of Finnish as a second language, there has hitherto been little research on the development of syntax. This article presents quantitative fi ndings of the use of a common and versatile syntactic pattern -the transitive construction -in texts written by adult learners of L2 Finnish, and rated on a functional CEFR scale at levels A1-C2. The aim is to examine how the use of the transitive construction develops across the levels: how it is varied and combined with diff erent kinds of syntactic resources, and moreover, what kinds of uses are typical at each level. The paper will focus on two clause level phenomena that also seem to indicate profi ciency in writing: the uses of the transitive construction with infi nite structures and passive. Language learning and linguistic development are discussed from the usage-based and Construction Grammarian perspective.

A.  Matasović on Balto-Slavic KS \ SK :
>
[fn] 12 A particular problem is posed by the equation of Lith. vãškas ‘wax’ with OCS voskŭ and OHG wahs.  If we start from *wokso-, then it seems that the RUKI-rule applied in Baltic before the metathesis of *ks > *sk, but not in Slavic.  On the other hand, the metathesis could be later than RUKI in both branches, and confirm that, at some period during the history of Baltic, *sk- was generalized word- initially, but *-šk- word-internally. Words showing word-internal *-sk-, such as Lith. druskà “salt”, would then have to be considered as younger formations, coined after that period.
>
In any case, Lith. kš instead of the expected ks is never quite reliable as an indicator of relative chronology, because it occurs in some very late borrowings, e.g. in Lith. krìkštas ‘baptism’ from OCS krĭstŭ.
>

Also see *H2awso-m ‘gold’, OLi. ausas, Li. áuksas (below, C).  I do not understand why he tries to separate the paths of vãškas & voskŭ, since Sl. *ṣk > sk instead of *ṣk > **xk would simply be failure of *ṣ > *x before K, in order to avoid KK.  It would be harder to explain if *Ks > *kš first (most now say that IIr. *š > S. ṣ, but I prefer IIr. *ṣ > Av. š, etc.).  Looking for full regularity here also seems impossible, since Li. has both -s- & -š- after RUKI, no known cause.  Also, the opposite metathesis (*SK > KS instead of *KS > SK) in :

*l(a)H2sk- > L. lascīvus, Sl. *laska ‘grace / love’, OCS laskati ‘to flatter’, R. laska ‘caress / kindness’, Li. lokšnùs ‘tender’

Hamp preferred an ending *-snu-, but this is certainly unneeded (and -nu- is seen elsewhere).  Though V_V vs. V_C could be part of the difference, there is still no full regularity.


B.  I think Uralic evidence also provides support for this sequence.  PU *wakša ‘wax’ is likely a loan << BS, & some see all PU *kš as evidence of the words containing it being loans.  This is because many ex. have clear matches in BS or IIr., with some ex. (Whalen 2025b) :

S. mákṣ-, mákṣā- ‘fly’, mákṣikā- ‘fly/bee’, Av. maxšī-, PU *mekše > Mv. mekš ‘bee’, F. mehi-läinen

*mH2ak- > OBg mokrŭ ‘damp / humid / wet’, LSb. mokšy ‘wet’, R. Mokša ‘a river’ >> Mh. mokša ‘a Moksha person’

PU *makša:y > *makša ‘rotten wood’, Mv. makšo, F. mahi, PU *mäkšä > EMr. mekš, WMr. mäkš

*wig^- ‘elm’ > OE wic, E. witch-elm, Al. vidh, Li. vìnkšna, PU *päkšnä > Es. pähn ‘elm / old lime tree’

and clearly related correspondences in :

*puk^syo- > Av. pusa-, NP fuš, S. púccha-m ‘tail / rod’, Hi. pūñch ‘tail / rear’, B. punzuṛɔ ‘tail’, Kva. pundzuṭɔ
PU *ponče ‘tail’ > Mr. pač, X. poč, Mi. ponš-pun ‘tailfeather’, Nen. panco ‘tail’, En. batu?o

S. muṣṭikā- ‘handful’, Ni. mustik ‘fist’, PU *mučkï ‘fist’ > Mv. mokšna, Mh. mokšenda, Mr.u. muškǝ̑ndo, Ud. mïžïk, Z. mïžïk


I also now think there is another :

Li. *blaH2sk-ti > blõkšti ‘to throw / fling’, bloškia ‘throws’, PU *pekše ‘arrow (with a blunt tip)’ > Mr.bk.u. pikš

Though this is not as obvious as the others, P()Vkš in both is too close to ignore.  Certainly not with so many other ex. being beyond chance (see C. for some ideas on specifics).  If these are also related, it would mean that every PU word with *kš had an IE equivalent with *Ks (or *sK in some, maybe also *Hs or *sH).  Thinking that, instead, all but one showed this match when blõkš- : *pekš- seems possible doesn’t seem likely to me.  This is more than a reasonable coincidence.  Also see C. for some IE *sK & *Ks > PU *śk & *kś > *ć near front V.


C.  For those who think maxšī- >> *mekše, another bug fits just as well :

Os. liskä, S. likṣā́, *linkṣī > A. liiṇṭṣií ‘nit’, PU *l’an’c’e ? > H. légy ‘fly’, Z. lödź ‘horsefly / gadfly’

but a loan from Dardic, the only branch with -n- and *-i:, would not follow any known history.  Changes like *i > *a are also indicative of a very old loan, if loan at all.


By this logic, it would be hard to separate OLi. ausas, Li. áuksas from PU *wäśkä > F. vaski ‘copper’.  However, there is no known way to make this work, and -ks- (and variant *-kš-) do not match *-śk-.  The presence of Ar. oski ‘gold’ & Su. guškin ‘gold’ also seem to complicate matters.  However, laryngeal metathesis was widespread in Indo-European (Whalen 2025a), so it would pay to examine oddities in roots with *H with this in mind.  For example, *H2aws- could also appears as *aH2ws- & *wesH2- in :

*H2awso-m > U. ausom, L. aurum ‘gold’, *aH2wso- > OLi. ausas, Li. áuksas, *wesH2-a: > *wesxa: > T. *w^äsa: > TA wäs ‘gold’, TB yasa

Here, H-metathesis is needed for the tone in *aH2wso- > Li. áuksas, for the *-e- in *wesHa: > T. *w^äsa:.  Adams has *-e- instead of **-i- since *wiso- > T. *wäse without *w^.  Since this *wesH2- indicates H-metathesis could happen before *H2e- > *H2a-, but many other IE have H-metathesis with no change to V, it must be a lasting optional change.  Compare also some *-e-H2- > *-aH2- in Celtic (Whalen 2025a).


To make some sense of all this, consider that *linkṣī > *l’an’c’e would require *i > *a with pal. of C.  If regular, *wesxa: > *w’asxa: would provide all the elements needed.  Metathesis *C’-C > *C-C’ would also exist in :

*mezg- > S. májjati ‘submerge/sink/dive’, Li. mazgóti ‘wash’, PU *mezg- > *miǝzg- > *m’ǝsk- > *mos’ke- ‘wash’

This set is well known as possible loans or cognates, so the changes required here (whatever the relation) should be applied to other words.  Also, it would show *i > *iǝ (as in Tocharian, the only branch with *wesxa: that can explain pal. in PU) to explain *CiV > *C’V and rounding of *ǝ > *o before other *ǝ merged with *a.  It can also combine with *H > k by s (Whalen 2024a) to make *sx > *sk :

*wesH2a:y > *wesxa:y > *wiǝsxä:y > PU *w’äskä > *wäs’kä > F. vaski ‘copper’

For opt. fronting, see *käktä \ *kakta.  It also had fem. *-a:y equivalent to most IE *-aH2-, also in (Hovers, Whalen 2025c) :

*kWetaH2- > R. četá ‘couple / pair’, SC čȅta ‘troop / squad’, Os. cäd(ä) ‘a pair of bulls in yoke’
*kWekWtaH2y- > PU *käktä \ *kakta


Also, Su. guškin ‘gold’ probably came from something like *wośki() to explain a loan >> Ar. oski ‘gold’ (maybe through intermediates).  This seems much too close to PU *wäs’kä < *wäs’kä:y to be chance, though I do not know what sound changes, etc., would be needed if cognate.


I think many other PU words match Tocharian ones, not just ‘copper’.  In Uralic, *mete ‘honey’ is supposedly a loan from IE, along with Ch. mì, J. mitsu, which most say << PT *miätu or similar.  I’ve mentioned many others (2025b, c, d) & in other papers on Uralic.  It is hard to think that so many PT features in PU could be due to ancient loans.  They seem to show that PU was a branch of IE, close to Tocharian.  This also fits with Hovers’ ideas on most PU words seeming to be from PIE, & I agree with about a 3rd of his cognates.


With this, *blaH2sk- > Li. blõkšti, PU *pekše would have to, by my rules (*wodo:r > *wodo:y > *wödöy > PU *wete), show something like *blaH2sk-on-s > *pa:sko:n > *paksoy > *pokšoy > *pökšöy > *pekše.  Rounding of *a > *o by *P seems to exist in PIE *(H2)appos > *oppyo > PU *ëppe > F. appi ‘father-in-law’, among others (forthc.).  The ending *yo is likely analogy < PU *äjjä ‘grandfather / old man’ (PIE *H2awyon- ‘uncle / grandfather’, etc.).  See Hovers for some other PIE *o > PU *ë when not changed by environmental causes (*st(o)rgo-s > Greek tórgos ‘vulture’, Gmc *sturkaz > English stork, Old Norse storkr, Hittite tarlā ‘stork’, PU *tërka ‘crane’).  If *app- \ *pap(p)- & *amm- \ *mam(m)- are seen as “natural” words in many languages throughout the world, PU having *ëppe instead of **appa would require some explanation anyway.

A.  There are many similarities between Uralic languages and PIE, including the reconstructed PIE lexicon.  Many of these have been considered loans before, and even such basic words as *wodo:r ‘water’ >> *wete > F. vesi have been proposed as loans.  I’m sure some loans certainly exist, but I want to consider the distribution of certain sounds to see which idea makes more sense.  The standard PU *š might have been *ṣ (Zhivlov), and it is found after *k in several words with IIr. kṣ & Balto-Slavic kš \ ks.  PU also had many *ks, and others don’t quite match, but a reasonable list of matches, loans or not :

S. mákṣ-, mákṣā- ‘fly’, mákṣikā- ‘fly/bee’, Av. maxšī-, PU *mekše > Mv. mekš ‘bee’, F. mehi-läinen

Os. liskä, S. likṣā́, *linkṣī > A. liiṇṭṣií ‘nit’, PU *l’an’c’e ? > H. légy ‘fly’, Z. lödź ‘horsefly / gadfly’

*maH2k- > Cz. mákati ‘make wet’, L. mācerāre ‘soften, make tender by soaking or steeping / weaken, waste away’
*mH2ak- > OBg mokrŭ ‘damp / humid / wet’, LSb. mokšy ‘wet’, R. Mokša ‘a river’ >> Mh. mokša ‘a Moksha person’
PU *makša:y > *makša ‘rotten wood’, Mv. makšo, F. mahi, PU *mäkšä > EMr. mekš, WMr. mäkš

*wig^- ‘elm’ > OE wic, E. witch-elm, Al. vidh, Li. vìnkšna, PU *päkšnä > Es. pähn ‘elm / old lime tree’

*puk^syo- > Av. pusa-, NP fuš, S. púccha-m ‘tail / rod’, Hi. pūñch ‘tail / rear’, B. punzuṛɔ ‘tail’, Kva. pundzuṭɔ
PU *ponče ‘tail’ > Mr. pač, X. poč, Mi. ponš-pun ‘tailfeather’, Nen. panco ‘tail’, En. batu?o


and also what would appear to be ṣṭik > *ṣṭk > *ṭṣk > *čk :

S. muṣṭikā- ‘handful’, Ni. mustik ‘fist’, PU *mučkï ‘fist’ > Mv. mokšna, Mh. mokšenda, Mr.u. muškǝ̑ndo, Ud. mïžïk, Z. mïžïk


and also *-ṣ > PU *-š :

*dhoiHnu-ṣ ? > S. dhenú- ‘giving milk’ >> PU *tejniš > Fc. *tiineš > F. tiine ‘pregnant [of animals]’, SEs. tiinõh

IIr. *vanaṣ- ? (1) ‘wood(en vessel)’ >> PU *weneš ‘boat’ > Kar. veneh, F. ven(h)e \ venho, Sm. *vënës > NSm. vanas, Mh. venež, Mv. venč


B.  Some of these must have been loans, the closer they were the more likely.  However, if Av. maxšī- >> PU *mekše, it would require *a-i > *e-i or similar, which did not happen later in PU, so it seems too old to be a loan.  If *wodo:r >> *wete, it would seem to require *-o:r > *-e & *o-e > *e-e, which would make *makše > *mekše more likely.  Again, no such changes seem to exist in PU, whatever the specifics.  If related, they would have to be much older loans or cognates.  These obviously seem related, so how?  Most IE words mean ‘fly’, only ‘bee’ in the east.  Also in the east is s > š after RUKI.  There is no reason for PU to have *kš instead of **ks if not due to RUKI.  This is not a change in PIE, but only one group.  Since no IE languages had *e in this word, this shows umlaut existed in PU, with other examples.  The ending *-iH2- ( > PU *-e (or *-i in other’s rec.)) is found only in Av. maxšī-, no other IE ending would cause umlaut, so why would *a-i > *e-e happen in a recent loan but not in native words?  With the needed elements, it would have to come from a protoform related to these words but with changes not found in any known IE donor.  Why would PU show such a mix of other IE forms & changes here if PU were not IE itself?  Since other IE have -u- (L. musca, maybe H. mušgalla- ‘caterpillar?’ ) it is unlikely it went back to PIE *a, maybe *mw- (Whalen 2025c).  It would be odd if PIE and PU were separate branches of a very old family but PU happened to show all the same changes of sound and meaning as in one sub-group of PIE.  If Uralic were not IE, there is no reason it would be associated with any of these forms or changes.


I say PU *makša:y > *makša \ *mäkšä to match *kWe-kWtaH2(y)- > PU *kakta:y > *käktä \ *kakta ‘2’, cognate with PIE *kWetaH2- > R. četá ‘couple / pair’, SC čȅta ‘troop / squad’, Os. cäd(ä) ‘a pair of bulls in yoke’ (Whalen 2025a).  Both have front & back variants, & I think this has to do with the PIE ending.  The Proto-Indo-European feminine of o-stems was *-o-iH2- > *-aH2(y)- (Whalen 2025b), with likely nom. *-aH2-s > *-a:H2.  My *-aH2(y)- explains TB -o and -ai-, among other retentions of -ai- & -ay- in other IE.  Some PU words that correspond to IE fem. have *-ä, others *-a (D).  If *kWe-kWtaH2(y)- > PU *kakta:y \ *kakta: > *käktä \ *kakta, it would help prove that *y existed here and was (one ?) cause of fronting in PU.  Since only Tocharian had many fem. with -ai-, a loan << BS doesn’t make sense for the V’s of *makša \ *mäkšä (unlike more recent Mokša >> mokša).  For ‘damp’ > ‘mold’ or ‘rotten wood’, see cognate L. mācerāre ‘weaken, waste away’ or other ex. (like PIE *mud- and other *muC-).


Though very similar, *linkṣī >> *l’an’c’e do not have the same V’s, and no process within PU could turn *i > *a (or whatever back *V existed, few Uralic cognates).  The *-i: > *-e (or *-i in some schemes) would match *mekše.  It would make sense if *i(:) palatalized both C’s, but then why not in *mekše?  Also, -N- only appears in one sub-branch of IE (IIr. > Dardic) :

*Hyork- > G. dórkai ‘eggs of lice/etc.’, *Hork- > Ar. ork‘iwn, *Hirk- > *rikH-? > Os. liskä, S. likṣā́, A. liiṇṭṣií ‘nit’

If it were a late loan from Dardic (or a very similar group), why would it not show the same changes as *mekše, which, if a loan, would also have to be recent & from a sub-branch of IE (IIr. > Ir.)?  If *kṣi > *kše but *ksi > *c’e, then only Baltic shows optional *ks > ks \ kš, etc.  The mix of features requires to explain just these 2 coherently as loans is hard to reconcile with known data, and there are more that show even greater mixes.


The -N- is also a problem in S. púccha-m, Hi. pūñch, PU *ponče ‘tail’.  The nasal is supposedly of Middle Indic age, so when would this supposed loan have taken place?  I’ve said (Whalen 2025d) that many of these changes were caused by Indo-Iranian nasal sonorants (r > n, y > ñ, w > m), seen often in Dardic and loans into nearby Burushaski, among others.  Even if old, is it likely that a word like IIr. *pućšỹa- would become PU *punčay > *ponče (or similar)?  No evidence of *u > *o existed in known PU changes.


For Li. vìnkšna, PU *päkšnä, the match is also of this type (with “extra” -N- and V’s not matching).  Though v- & p- do not match, there is no reason why *v > *b > p would not work, if old, loan or not.  This might have happened in old forms of PU or in the IE donor (Tocharian alone had some old *w > p, *p > w, no regularity).  Why would such a cluster as kšn exist in both if unrelated?  *kšn is rare in PU, and some say it was from *kšVn, which would not work if related to vìnkšna.  If they’re related and old, only Baltic has kšn in this word among IE, so it would be useless for a “long-range” comparison.  The creation of retroflex after RUKI only happened in a subset of IE, so the same change in PU would be unexplained if not IE itself.  The same in *mekše, *makša:y, etc.


For nkšn vs. *kšn, other Baltic words show *-KSN- > -NKSN- :

*pluHksmāH2, Li. plū́ksna \ plù(n)ksna ‘feather, quill’, L. plūma ‘feather, plume’

Li. ū́kas ‘fog’, ū͂kti ‘get dark/foggy’, ūksmė͂ \ uñksnė \ unksnė͂ ‘shade / shady place’, ūksnė͂ ‘shade’

Again, this loan would have to be older than attested Baltic forms, which would not matter if a loan from another IE branch.  However, since vìnkšna is probably analogical after ãlksnis and other trees with -ksn- / -kšn- :

*Halsno- > L. alnus, Li. ãlksnis ‘alder’, élksna \ álksna ‘alder thicket / marsh’

Li. šermùkšnis / -nė / -lė ‘mountain ash’

*bhrHg^ó- ‘birch’, *bhrHg^isno- > *frākhisno- > L. frāxinus \ *fārksnos > farnus ‘ash’

no PIE word with *-ksn- for ‘elm’ would exist.  Only Tocharian If Tocharian (if *w > p there) were very distant from other IE, then the match would require a loan from Baltic (since adding -kšn- is a very late change not even seen in Slavic), but with the change w > p.  What sequence of events would allow PU speakers to move from the lands near the Baltic Sea to central Asia and back again to get both BS & Tocharian features in one word?  And why pick up these specific loans and changes?  Nothing but a long association with one or both groups makes sense.  Since most Tocharian words for types of trees are unknown, if it was shared with Baltic it would require part of this to take place in East Europe.  The best sequence for this and other data is that Tocharian was a fairly normal IE language, with sound changes shared by many other close IE groups.  PU *päkšnä would have to be from a branch of IE, and I see no reason that Uralic would not be a branch of IE.  Other changes show Uralic was either a sub-group of Tocharian or very, very similar to it.


Baltic also seems to alternate ksn / ksl / gzd with no cause.  In addition to šermùkšnis / -nė / -lė, see :

*g^hwoigW- > G. phoîbos ‘pure / bright’
*g^hwoigW-zCa: > Li. žvaigzdė, Lt. zvaigzne ‘star’
*gWhwoigW-zCa: > Slavic *gwaigzda: > Po. gwiazda

The same would then need to exist in :

*wig^- ‘elm’ > Os. wis-qäd ‘maple’, *wikšna: > *wikšla: > *wikštla: ? > PU *wakštera ‘maple’ > Mr. waštar, F. vaahtera

If a loan from Baltic, consider that *i > *a in 2 words for trees would show a change, but not a recent or known one.  This has been seen as a loan, but, of whatever source, it would still show a change *kšn/t/l or similar, maybe related to Baltic ones but different.  With several words showing KS but not with other features close enough to be loans, where does the need for IE words in PU to be loans exist?


C.  There are other words that make it clear that *r also sometimes caused ret., even at a distance, just as in Indic :

*ser- ‘flow’, *seraH2- > PU *sara \ *šara ‘flood’ > Mi. *tūr, X. *Lār, Hn. ár

If not, the differing C- would have no cause.  A ret. *ṛ in PU would be too close to that in several IE branches to be chance, especially when RUKI in *ks > *kš seems needed.


That PU *čr existed is seen in cognates with *č vs. *r.  This internal evidence is enough for PU, and the words they exist in have clear IE cognates, like Fc. *š(r)äšnä > F. hähnä, Es. rähn (bel...

Si rimanda alle più significative recensioni del volume:
- Franca Bellucci, Dialoghi mediterranei, 01/01/2021 | «Kavafis ovvero il faro di luce nel Mediterraneo. Divagazioni»
- Paolo Rigo, Scaffale Aperto. Rivista di Italianistica - Anno 11 (2020), 01/12/2020 | «Tutte le poesie di Kavafis»
- Angelo Molica Franco, Il Fatto Quotidiano, 03/04/2020 | «Eterno Kavafis: il futuro di tutti è nel passato»
- Matteo Marchesini, Il Foglio, 01/04/2020 | «Kavafis e la cornice del suo presente assoluto»
- Livio Partiti, Il posto delle parole, 21/03/2020 | «Intervista a Paola Maria Minucci»
- Alessandra Pacelli, Il Mattino, 11/03/2020 | «Tra "Itaca", inediti e testi segreti. Tutto Kavafis, il poeta del futuro»
- Matteo Marchesini, Radio Radicale, 06/03/2020 | «Critica e militanti»
- Massimo Natale, Alias - Il Manifesto, 16/02/2020 | «Kavafis - Tutte le poesie tradotte da Donzelli»
_ Miska Ruggeri, Mizar - Tg 2, 15/02/2020 | «Disseppellire i versi di Kostandinos Kavafis»
- Luca Mastrantonio, Sette, 07/02/2020 | «L'Ulisse di Kavafis e l'amore di Jacqueline Onassis»
- Globalist, 01/02/2020 | «Tradotto tutto Kavafis, anche le poesie più segrete»
- Radio 3 Suite, 01/02/2020 | «Intervista a Paola Minucci»
- Remo Curi, La Gazzetta di Parma, 29/01/2020 | «Tutte le poesie, inno all'ascetismo aristocratico della lirica»
- Massimo Raffaeli, Venerdì di Repubblica, 24/01/2020 | «La grande bellezza di Kavafis»
- Gazzetta del Mezzogiorno, 20/01/2020 | «Kavafis oltre la sua "Itaca", una raccolta di tutte le poesie»
- Francesca Valenti, Libero, 19/01/2020 | «I versi rifiutati e segreti in un testo. Tutte le poesie di Konstandinos Kavafis»
- Massimo Onofri, Avvenire, 18/01/2020 | «Fra Grecia e Cristianesimo s'invola il verso di Kavafis»
- Andrea Galgano, Frontiera di pagine, 14/01/2020 | «Konstantinos Kavafis: l'intuizione cosciente»
- Roberto Galaverni, La Lettura, 29/12/2019 | «Mito Kavafis, la voce più nascosta»
- Franco Marcoaldi, Robinson, 28/12/2019 | «L'inattualità sempre attuale di Kavafis»
- Davide Brullo, Il Giornale, 27/12/2019 | «"Tutte le poesie" di Kavafis sono un inno all'antichità»

[ https://www.donzelli.it/libro/9788868439965 ]

Summary in North Sámi of my research on the Sámi yoik (2014-2015)

Dán teavsttas leat guokte ulbmila. Ovtta dáfus dat lea bajilgovva mu dutkanbohtosiin ja dahká daid olahahttin dárogielat ja sámegielat olbmuide, ležžet dál studeanttat, juoigit dahje dušše sáhkkes lohkkit. Nuppe dáfus lea ulbmil govvidit juoigama «fámolaš» vierrun. Dat mearkkaša ahte luohti ii leat dušše čáppa «álbmotmusihkka». Juoigan lea maid vuohki movt áŋgiruššat máilmmis. Dat lea fápmu mii addá erenoamáš vásáhusaid mat, olu juigiid mielas, olahit «guhkkelii go sánit». Váldogažaldat teavsttas leage ná: Makkár oktavuođa oažžu birrasiidda juoigama bokte? Nuppiiguin sániiguin, movt sáhttá luohti váikkuhit olbmo oktavuhtii eará olbmuide ja servodahkii, dahje elliide ja báikkiide?

Direct object clitics in Modern Standard Romanian display different properties depending on whether or not they double an object. We propose a dual analysis for the clitics: they function as agreement makers when they double an object and as pronouns when they do not. Furthermore, the lexical entries differ beyond the presence or absence of pronominal referential features, and this accounts for the split behavior. The analysis is placed in its historical context and extended to other varieties of Romanian. Finally, we argue that the Romanian lexical split is not an isolated phenomenon: multiple similar splits can be found in the typology of agreement marking.

The Finnish Literature Society (SKS) was founded in 1831 and has, from the very beginning, engaged in publishing operations. It nowadays publishes literature in the fields of ethnology and folkloristics, linguistics, literary research and cultural history. The first volume of the Studia Fennica series appeared in 1933. Since 1992, the series has been divided into three thematic subseries: Ethnologica, Folkloristica and Linguistica. Two additional subseries were formed in 2002, Historica and Litteraria. The subseries Anthropologica was formed in 2007. In addition to its publishing activities, the Finnish Literature Society maintains research activities and infrastructures, an archive containing folklore and literary collections, a research library and promotes Finnish literature abroad. Studia fennica editorial board Markku Haakana, professor

...A thing that is not globally shared, though, is the languages we speak. We speak languages, distinct or related, as a form of communication. Mostly, we have different words for the same things in different languages. Yet some words do not exclusively belong to a specific language, such as the word “coffee”. This makes coffee fairly interesting, as the word is far too common across high amounts of languages for it to be a coincidence. This essay will explore the typology of the word “coffee” and the history that led to its globalisation in support of validating the linguistic commodity of the word coffee as a global bridge.

In this paper I reflect on methodological aspects to take into consideration when developing methods for investigating three- to five-year-old children’s beliefs about language, language use and bilingualism. I analyze participant observations and notes taken in the field. The study focuses on bilingual Finnish-Swedish children in Swedish-medium preschools in Finland. At one of the schools, most of the children and I did not share a common language, so the interaction between us heightened both the children’s language awareness and that of my own. This drew my attention to communicative aspects of embodiment and multimodality and to the distribution of responsibility for interaction. I detected two interaction orders in which children’s agency stood out in their intention to make their voice heard, and I used my experiences to develop an ethically-oriented data-generating approach to enhance communication about communication.

Державний заклад «Південноукраїнський національний педагогічний університет імені К. Д. Ушинського» «Складівниця української мови»: словник складів від О. Різниківа Васильєва К. П. «Складівниця української мови»: словник складів від О. Різниківа. У статті досліджується словник складів О. Різниківа «Складівниця української мови», який ще не ставав об"єктом спеціального розгляду. Зазначається, що О. Різників перший здійснив спробу зібрати, виділити та впорядкувати найхарактерніші склади для української мови, яких виявилося близько 9600. Підкреслюється, що автор словника пропонує ввести в мовознавчу науку термін «слов"як», що означає склад чи групу складів осмисленої мови. У статті завдяки розподілу всіх складів за схемами було визначено кількість різних можливих форм складів, багатство яких підкреслює особливість української мови, сполучуваність, валентність голосних і приголосних звуків, обґрунтовується виділення в окрему групу складів, що мають лише приголосні. Ключові слова: словники, «Складівниця української мови», склад, слов'як, голосні та приголосні звуки. Васильева Е. П. «Слоговница украинского языка»: словарь слогов от А. Ризныкива. В статье исследуется словарь слогов А. Ризныкива «Слоговница украинского языка», который еще не становился объектом специального рассмотрения. Отмечается, что А. Ризныкив первый предпринял попытку собрать, выделить и упорядочить характерные слоги для украинского языка, которых оказалось около 9600. Подчеркивается, что автор словаря предлагает ввести в языковедческую науку термин «словьяк», что означает слог или группу слогов осмысленной речи. В статье благодаря распределению всех слогов по схемам, было определено количество различных возможных форм слогов, богатство которых подчеркивает особенность украинского языка, сочетаемость, валентность гласных и согласных звуков, обосновывается выделение в отдельную группу слогов, имеющих только согласные. Ключевые слова: словари, «Слоговница украинского языка», слог, словьяк, гласные и согласные звуки. Vasylieva K. P. «The Ukrainian language syllables book»: the dictionary of syllables by O. Riznykiv. The author of the article examines the dictionary of syllables by O. Riznykiv «The Ukrainian language syllables book» and determines the uniqueness of such a work, that has not been the object of special consideration yet. It is noted that O. Riznykiv, using the rules of syllabification was the first who made an attempt to collect, identify and systematize the most typical syllables characterized for the Ukrainian language. The number of them approaches approximately 9.600, and the dictionary covers 97-98% syllables from the Ukrainian literary language. It is emphasized that the author of the dictionary proposes to introduce into the science of linguistics the Ukrainian term «slovjak», which means a syllable or a set of syllables of a meaningful speech. In order to interest the readers, the dictionary by O. Riznykiv «The Ukrainian language syllables book» also contains the examples concerning «the mysterious nature of speeches», gives the linguists and theologians" quotations and tries to reveal the links of speech with breathing. Due to the division of all syllables into the schemes, the article determines the number of different possible forms of syllables and its great diversity underlines the peculiarity of the Ukrainian language. It also defines the compatibility, the arrangement of vowels and consonants and justifies the allocation of a separate group of syllables that have only consonants.

Valtonen, T. (2024). Miksi keminsaame kuoli? : Sosiolingvistinen rekonstruktio keminsaamen sammumiseen johtaneista tekijöistä. Suomalais-Ugrilaisen Seuran Aikakauskirja, 2024(100), 207–239. https://doi.org/10.33340/susa.140998

Kemi Saami is an extinct language that was spoken in the administrative region of Kemi Lapmark in present-day Northern Lapland, Finland. This article analyzes the sociolinguistic processes that resulted in the extinction of Kemi Saami. The analysis is based on contemporary descriptions from the eighteenth and nineteenth centuries. The information preserved in the descriptions is analyzed using the sociolinguistic analysis template developed under the European Language Diversity for All (ELDIA) project, as well as UNESCO's nine factors describing the situation of an endangered language. The theoretical background is based on Joshua Fishman's work on language shift. The results show that the extinction of Kemi Saami was a consequence of the language's weak social status, the church's language policy, and several social and cultural dislocations that the language community experienced. The most significant of these dislocations was the arrival of Finnish settlers and the agrarian socioeconomic model. The resulting population growth and environmental changes forced the inhabitants to give up the old way of life based on hunting and fishing. This led further to the deterioration of the socioeconomic situation of the language community in relation to the settlers and to the rapid weakening of cultural self-esteem. As a result, the members of the language community began to be ashamed of their own language and culture and wanted to give them up for Finnish language and culture.

SKY Journal of Linguistics 19 (2006), 131–146 Anna Fenyvesi and Gyula Zsigri The Role of Perception in Loanword Adaptation: The Fate of Initial Unstressed Syllables in American Finnish and American Hungarian* Abstract This paper examines one aspect of the phonological ...

Results from a recent taxometric investigation of the alexithymia construct, measured by the 20-Item Toronto Alexithymia Scale (TAS-20), with English-speaking samples in Canada provided evidence that alexithymia is best conceptualized as a dimensional rather than a categorical construct. The aim of the current investigation was to attempt to generalize the outcome of this earlier investigation by examining the latent structure of the alexithymia construct in a sample whose primary language was not English. A second aim was to examine the latent structure in men and women separately as some investigators propose that typological differences in the nature of alexithymia may differ by gender. The sample comprised 5194 (2377 men, 2817 women) residents of Finland who as part of a population-based health survey, supported by the Finnish government, had completed a Finnish-language translation of the TAS-20. Three non-redundant taxometric analyses were performed in the total sample and for men and women separately, using item sets from the TAS-20 as indicators. The outcomes of the taxometric analyses for all three samples were similar to those in the Canadian study. These findings provide further evidence that alexithymia is a dimensional construct.

Kirkkomuusikoiden seurakuntatyössä on monia äänenkäytöllisiä haasteita ja riskitekijöitä. Kyselytutkimuksella selvitin kanttoreilta, mitä äänellisiä riskejä heillä on omasta mielestään työelämässä, kuinka he hoitavat ääntään ja millaisia ääniongelmia heillä on ollut. Työssäni käyn läpi myös lauluääneen vaikuttavia anatomisia rakenteita ja klassisen äänenmuodostuksen perusperiaatteita. Esittelen myös yleisimpiä ääniongelmia seikkaperäisesti

German, Dutch, and English nickname formation is examined using a contrastive corpus of nicknames which were found in the online profiles of amateur athletes and are compared with the same individuals’ first and last names. We study the word formation and word creation of nicknames, either based on the athletes’ legal names or coined freely, pointing out parallels and divergences between the three languages. Two prototypes are identified crosslinguistically as relevant bases for output schemas: disyllabic trochees ending in -i (cf. German Conni, Dutch Passie, English Thanny) and monosyllabics ending in a closed syllable containing a single sonorant (Sash, Bous, Maze). These structures are then interpreted in terms of preferred sound patterns and sex marking. Dutch turns out in many respects to hold an intermediate position between German and English.