Perspectives

of Slavonic
Etymology

Edited by
Bohumil Vykypěl
and Vít Boček

Nakladatelství
Lidové noviny
Praha 2016
The present volume was prepared with the support of a grant from the
Czech Science Foundation “Etymological Dictionary of the Old Church
Slavonic Language: Summing up a Generation Project” (Nr. 13-17435S).

Edited by Bohumil Vykypěl and Vít Boček

Reviewed by Ljubovʼ V. Kurkina and Ľubor Králik

Studia etymologica Brunensia 19

Eds. Ilona Janyšková & Helena Karlíková

© editors

ISBN 978-80-7422-491-1
Table of contents

Table of contents / 5
By way of introduction / 7

Ilona Janyšková (Brno)

О будущем славянской
этимологической лексикографии / 9

Vít Boček (Brno)

On tools for the classification and evaluation
of (Slavonic) etymological dictionaries / 19

Metka Furlan (Ljubljana)

Slavic etymology between linguistic
predictability and unpredictability / 33

Georg Holzer (Wien)

Slavische Etymologie
und generative Phonologie / 49

Hristina Deykova (Sofia)

Probleme der bulgarischen Etymologie.
Slawisch-balkanische Sprachinterferenz
(Aus der Praxis des „Bulgarischen
etymologischen Wörterbuchs“) / 65

Сергей А. Мызников (Санкт-Петербург)

Этимологические исследования
русской диалектной лексики:
исконные и заимствованные пласты / 75
Bohumil Vykypěl (Brno)
Marginalia etymologica slavica ii / 83

Александар Лома (Београд)

Из работы над этимологическим
словарем сербского языка.
Этимологии вверх дном / 89

Ranko Matasović (Zagreb)

Seven Slavic etymologies:
*ędro, *gъrdъ, *koristь, *kosъ,
*myslь, *naglъ, *němъ / 105

Robert Orr (Ottawa)

Slavic morphology in Tolkien?
Sméagol and Ungol / 113
ranko matasović: seven slavic etymologies: 105
*ędro, *gъrdъ, *koristь, *kosъ, *myslь, Matasović
*naglъ, *němъ

This paper offers seven new etymologies of Proto-Slavic words for which no
convincing etymology has been found so far. Although they are speculative, we
believe they represent valid and viable alternatives to the etymologies found
in reference books.

1 *ędro ‘kernel’

Julius Pokorny (IEW 774) derived both PSl. *ędro ‘kernel, core’ (Russ. jadró ‘ker-
nel, core’, Cz. jádro ‘kernel, core’, Pl. jądro ‘grain, kernel, core’, Croat. jédro ‘id.’,¹
Bulg. jedró ‘kernel’) and *ědro ‘bosom’ (OCS jadra pl. ‘bosom, embarce’, Cz. ňadro
‘breast, bosom’, Pl. jadro ‘net’, Croat. jȅdro ‘sail’, njȅdro ‘bosom’)² from the same
PIE root, *h₂eyd- ‘swell’ (Gr. oidéō ‘swell’, Arm. ait ‘cheek’, aitnum ‘swell’).³ How-
ever, this is difficult from the phonological point of view, and it certainly does
not suffice to claim, as IEW does, that *ędro is from a ‘nasalised’ form of the
root. Before accepting such an etymology, we would need to show the source
of the nasal in *ędro, and also to explain how the meanings ‘kernel, core’ and
‘bosom’ are related. No wonder, then, that Pokorny’s etymology is rejected by
most contemporary compendia of Slavic etymology (Derksen 2008: 150, 157,
ESSJa 6: 45–47, 65–66, Vasmer 1953–1958, 3: 482–483). Yet rejecting the connec-
tion does not solve the problem, as it leaves PSl. *ędro without an etymology.
Therefore, in this paper, we shall attempt to show that Pokorny was right after
all, and derive both *ędro and *ědro from the same PIE source.
Here is how we proceed. First, we note that both ‘bosom’ and ‘kernel, core’
can be derived from the original meaning ‘internal organ, gland’. Next, we
recall that many nouns denoting internal organs in PIE were heteroclita, hav-
ing the alternation of *-r (in the Nom./Acc. sg.) and *-n- (in the oblique cases),

1 Croat. jédro is slightly archaic; in the contemporary language it has been all but replaced
by jézgra ‘kernel, core’, with the cluster -zgr- by analogy with mezgra ‘core, sap of a tree’.
2 The forms with initial n- are from the syntagm *vъn ědra ‘in(to) the bosom’. It is not quite
clear whether PSl. *ędrъ ‘strong, firm, solid’ (OCS ędri ‘quick’, Croat. jédar ‘firm, solid’,
Bulg. édăr ‘big, strong’, Derksen 2008: 157) belongs together with *ędro, but this is possible
if one assumes the semantic development from ‘having a strong core’ to ‘firm, solid’ and
‘strong’.
3 Armenian vocalism shows that the initial laryngeal was *h₂- rather than *h₃-. The Greek
form is explainable by positing a PIE intensive or causative *h₂oyd-eye-, cf. LIV 230.
 106 e.g. *yēkwr ‘liver’ (Gr. hē͂par, hḗpatos, Lat. iēcur, iecinis, etc.). Heteroclita were
Matasović not preserved as such in Slavic, but there is some clear evidence that they still
existed in the Balto-Slavic period (Matasović 1998).
We posit a PIE heterocliton *h₂eydr/*h₂idn-os, which would have given
Balto-Slavic *aydr/*idn-as. In the oblique cases, *-dn- would have regularly
metathesized to *-nd- (as in Lith. vanduo͂ ‘water’ < *wodn-), so we have to posit
the stems *aydr and *inda-. While they were still parts of the same nominal
paradigm, the Nom./Acc. stem influenced the stem of the oblique cases, which
was re-shaped as *indra-, and this was then the origin of the separate noun
meaning ‘kernel, core’ (> PSl. *ędro), while the stem *aydr was thematized to
*aydra- and became the origin of the noun meaning ‘swelling’ > ‘bosom, breast’
(> PSl. *ědro). The name of the Vedic god Indra-, although it is sometimes con-
nected with these Slavic words, is probably unrelated.
The next question to address is whether there is any independent evidence
for a PIE heterocliton meaning, roughly, ‘a swelling’, ‘internal organ’, and/or
‘a gland’. Indeed, there is: in a number of IE languages we find derivatives
of this root, built with suffixes *-rV- and *-nV-, which are best explained as
generalizations of one stem of an original PIE heterocliton (cf., e.g., ON fýr
‘fire’ and Goth. fōn ‘id.’, both from PIE *peh₂wr/*puh₂n-os ‘fire’, Hitt. paḫḫur,
peḫwenaš). These derivatives include OHG eitar ‘poison’ < PGerm. *aitra-, as
well as Lith. jedrà ‘measles’ (probably originally ‘swellings’),⁴ which are parallel
to PSl. *ědro. A derivative from the same root with the suffix *-n- is OIr. inne
‘quality, manner, meaning’, but more concretely also ‘heart, bowels, interior
parts’, which probably represents the original meaning. In the case of Ger-
manic *aitra- we assume that the meaning ‘poison’ developed from ‘excretions
of glands’, cf. a similar development in PSl. *jadъ ‘poison, gall, anger’ (OCS jadъ,
Croat. jȁd, Russ. jad, Cz. jed, Pl. jad), which is from the same PIE root *h₂eyd-.
On the other hand, OIr. inne can be regularly derived from PCelt. *inniyā <
*idnyā, from PIE *h₂id-n-, with the generalized stem of the oblique cases of the
heterocliton *h₂eydr/*h₂idn-os.
It is at present unclear whether the Baltic river-names mentioned by Po-
korny (Indus, Indura, Indra, etc., IEW 774) are indeed from the same root. Latv.
idra ‘faule Mark eines Baumes’ probably reflects a hybrid *idrā < *h₂idreh₂, but
the absence of lengthening by Winter’s law (we would expect *īdra) is puz-
zling. The original accentuation of *ědro is difficult to ascertain, and Derksen
(2008: 150) does not reconstruct it. The Croatian dialect forms do not agree:

4 The form jedrà is probably originally a dialectal variant of *iedra with *ie regularly from
BSl. *ay < PIE *h₂ey. Such by-forms are well attested in Lithuanian, e.g. Lith. dial. jeva
‘Prunus padus’ besides (Standard) ieva (Illič-Svityč 1963: 68).
jȅdro (Standard Štokavian), jȁdro (Novi), but jȃdro (Grobnik). Original accent 107
paradigm (henceforth a.p.) c cannot be excluded. PSl. *ędro, on the other hand, Matasović
belonged to a.p. b (cf. Russ. jadró, Croat. jédro). This means that the proto-form
did not undergo Winter’s law, which was blocked by a nasal.

2 PSl. *gъrdъ

PSl. *gъrdъ (a.p. c) is reconstructed on the basis of OCS grъdъ ‘haughty’, Slov.
gȓd ‘ugly’, Russ. górdyj ‘proud’, Cz. hrdý ‘proud’, Pol. dial. gardy, etc. (cf. Derksen
2008: 198, ESSJa 7: 206f.). This word is traditionally derived from PIE *gwerd-
(Lith. gurdùs ‘weak, slow’, Latv. gur̃ ds ‘tired’ and Gr. bradýs ‘slow’ (cf. Vasmer
1953–1958, 1: 294–295, Smoczyński 2007: 212). Lat. gurdus ‘idiot’ is also some-
times connected, under the assumption that it originally meant ‘slow’ (De Vaan
2008: 275). However, the difference in meaning between these words is consid-
erable. If this etymology is accepted, one would have to assume the semantic
development from ‘slow’ to ‘rigid’ and ‘haughty’, for which we have no exact
parallels. Just as unconvincing is the derivation of PSl. *gъrdъ from PIE *gwerH-
‘heavy’ (Skr. gurú-, Gr. barýs, Lat. gravis, etc., IEW 476) with the suffix *-dh- and
the semantic development from ‘heavy’ to ‘rigid’ and ‘proud’ (Snoj 2003: 189).
I would like to propose a new etymology, which has the advantage of posit-
ing a less complex semantic development in Slavic. I believe that PSl. *gъrdъ is
an old compound, PIE *gwrH-dhh₁-o-, containing the root *gwerH- ‘praise’ (Skr.
gir-, Lat. grātus ‘thankful’, etc., IEW 478) and the verbal root *dheh₁- ‘do, make’
(Lat. fēcī, Skr. dhā-, OCS děti, etc., IEW 235ff.). A parallel can be found in PCelt.
*bardos ‘bard, a poet composing eulogies’ (Gaul. bardos, Welsh bardd, etc.) < PIE
*gwrH-dhh₁o-, cf. Matasović (2009: 56) and the Skr. phrase gíras dhā- ‘praise’,
litteraly ‘to make praises (gir)’. The semantic development was from ‘making
praises’ to ‘haughty, prepotent’ and ‘proud’.⁵ The formal side of this derivation
is straightforward, as PIE *C₁rC₂ yielded *C₁urC₂ when C₁ was a velar.

3 PSl. *kòristь

PSl. *kòristь is reconstructed on the basis of the following correspondences:

OCS koristь τὰ σκῦλα ‘booty’, Croat. kȍrīst, Slov. korȋst, Pol. korzyść ‘profit,
earnings’, Cz. kořist ‘booty’, Ukr. korýst’ ‘use, profit, advantage’, Russ. korýst’
(borrowed from Ukrainian because of -y-). Besides the noun, many Slavic

5 There is no reason to assume, as Young (2006) does, that the Slavic word is a cognate of
Lith. participle pagìrtas ‘praised’ (from pagìrti ‘praise’), as the acute in Lithuanian does
not correspond to the circumflex in Slavic.
 108 languages also have the denominative verb *korìstiti/*korìstati (Croat. kòristiti,
Matasović Slov. korístiti, Russ. korýstit’sja ‘envy’, Ukr. korystáti, Pol. korzystać, earlier
korzyścić ‘use’). The earliest meaning of PSl. *kòristь may have been either
‘booty’ (as in OCS) or ‘profit’ (found in most Slavic languages).
This word does not have a convincing etymology. Skok (1985: 155) thought
that the derivation from *koriti ‘reproach, scold’ was most probable, under
the assumption that the verb originally meant ‘to win a battle’. Even if that
were correct, the word-formation of *koristь would still remain unexplained.
Therefore Vasmer (1953–1958, 1: 638) disputed that etymology, as well as the
connection with the noun *kora ‘bark’ (suggested by Brückner), but he did not
offer any alternative.
Snoj (2003: 310, developing an idea by Bojan Čop) suggests that *koristь is an
old compound of *korih₂- ‘army’ (From the root of Lith. kãras ‘war’) and *dhh₁-ti-
(from the root *dheh₁- ‘do’ > OCS děti, Lith. dė́ti, etc.). The original meaning would
have been ‘that which has been placed to the disposal of the army’ > ‘booty’, but
this is not very convincing. ESSJa (11: 71–72, cf. also Gluhak 1992: 331–332, Varbot
1972) adduces a thoroughly improbable etymology according to which the verb
*koristiti/*koristati was formed with the ‘expressive’ prefix *ko- and the root of
the noun *ristь ‘grasp’ (cf. Russ. ristát’ ‘run’, OCS ristati, Lith. rìstas ‘quick’). The
existence of the prefix *ko- in Proto-Slavic is very doubtful (Matasović 2014),
so this etymology does not hold water. All of this allows us to conclude that no
truly convincing etymology of *koristь has been suggested so far.
I believe that *kòristь is an old loanword from Vulgar Latin that spread by
trade, especially in the language of merchants on the Danube in the 6th cen-
tury A.D. The source is VLat. caristia ‘rise of price’ (REW 1694a, p. 161). This
Vulgar Latin word is reflected in, e.g., It. carestia, Sp. carestía and Port. carestia.
In Romance languages, the attested meanings are usually ‘rise of price’ and
‘pricyness’, sometimes also ‘poverty’ (in Spanish and in Italian).
The phonological adaptation of VLat. caristia to PSl. *kòristь is straightfor-
ward. We only need to assume that the original feminine noun in –ia became an
i-stem in Slavic, rather than a jā-stem. The replacement of VLat. -a- by Slavic
*o is regular, cf. Lat. lactūca ‘lettuce’ > Croat. lòćika, Lat. Salōna > Croat. Sòlīn.
Although we would expect the short Latin -i- to be reflected as PSl. *ь rather
than *i (cf. VLat. missa ‘mass’ > *mьša > Croat. dial. maša), there are exceptions to
this rule, e.g., Lat. viscum ‘mistletoe’ > Croat. dial. biska, Lat. canistrum ‘basket’ >
Croat. dial. kònistra (attested since the 15th century, cf. Skok 1985: 144).
On the semantic side, we have to assume that the original meaning in Slavic
was ‘merchandise’, which could have easily developed from ‘pricyness’, ‘rise of
prices’ in the language of merchants. Thence, the semantic development led
to the attested meanings ‘booty’ (< *‘robbed goods’) and ‘use, profit’ (< *‘usable 109
goods’). Matasović
The origin of VLat. caristia is uncertain. It has been argued that it is a de-
rivative from the adjective carus ‘dear, expensive’, or from the verb carēre ‘lack’.
If the latter is true, the other meaning of caristia was ‘poverty’.

4 PSl. *kòsъ

PSl. *kòsъ (a.p. c) ‘blackbird, Turdus merula’ is reconstructed on the basis of

Croat. kȏs, Bulg. kos, Russ. dial. kos, Ukr. kis, Cz. kos, Pol. kos, etc. This word
has been connected with Gr. kópsikhos, kóssyphos ‘blackbird’ (Derksen 2008:
239, ESSJa 11: 175ff., Skok 1985: 160, Gluhak 1992: 335, cf. also Matasović 2013).
Beekes (2010: 760) doubts this etymology and considers the Greek words as
loanwords from some substratum (“Pre-Greek”). He also rejects the idea that
both the Greek and the Slavic words for ‘blackbird’ are derived from *kop-so-,
where *kop- is the root ‘to dig’ (OCS kopati, Lith. kapóti, Gr. kóptō ‘strike’) on
the grounds that *-so- is not a nominal suffix. However, it is possible to de-
rive PSl. *kosъ from a nomen agentis *kosó-, from the root *kes- ‘comb, scratch’
(OCS česati, Hitt. kišzi, Lith. kasù, kàsti ‘dig’, kasaũ, kasýti ‘scratch’, IEW 585). As
blackbird is a bird that is known to scratch the ground in looking for food, it
would not be surprising that its name is, litterally, ‘scratcher’ in Slavic. For the
semantic motivation cf. Croat. češljugar ‘European goldfinch, Carduelis cardu-
elis’ (from češljuga ‘comb’).

5 PSl. *myslь

PSl. *myslь ‘thought’ (a.p. a or a.p. c) is reconstructed on the basis of OCS myslь,
Slov. mȋsel, Croat. mȋsao, Russ. mýsl’, Pol. myśl, Cz. mysl (cf. ESSJa 21: 47–50). This
word has been traditionally connected with PIE *mewdh- ‘beware, think’ (Lith.
maudžiù, maũsti ‘ache, long for’, Goth. ga-maudjan ‘remember’ (cf. Skok 1985:
431–432, Vasmer 1953–1958, 2: 184–185), but this leaves Slavic *-y- unaccounted
for. Another possibility is to relate *myslь to Gr. mỹthos ‘story, talk, word’ and
derive both words from PIE *muHdh- (Derksen 2008: 337, ESSJa 21: 47–50), but
this implies that we have an exclusive Greek-Slavic lexical isogloss, and the re-
flexes of the root *muHdh- in other languages remain unknown. Beekes (2010:
976) remains sceptical and does not even mention this etymology. From the
point of view of word-formation, mỹthos is isolated in Greek, and it may well
be a loanword from some unknown source. A third possibility would be to
connect *myslь and Lat. moveō ‘move’ < *m(y)ewH- (cf. also Skr. mī́vati, Lith.
 110 máuju ‘put on’, IEW 743). Although *myslь would be regularly derivable from
Matasović *muH-sli-, the difference in meaning is considerable, and the complex seman-
tic evolution that would need to be assumed (? *‘motion’ > ‘act’ > ‘thought’)
renders this etymology improbable.
As no other convincing etymology is available, I would like to propose that
*myslь is derived from the root *mewd- ‘rejoice’ (Skr. módate ‘rejoices’, IEW
741–742) with the suffix *-sl- or *-dhl- (cf. Matasović 2014: 44, 106f.), with the se-
mantic development ‘joy’ > ‘joyful thought’ > ‘thought’. The long vowel in Slavic
would be expected because of Winter’s law, and the acute (probably implied
by the accentuation of Russ mýsl’, Gen. sg. mýsli) would be regular. However,
this renders the connection with Lith. mudrùs ‘lively’ difficult, as the operation
of Winter’s law would have yielded *mū́drus. At least formally, Lith. mudrùs is
better derived from PIE *mewdh- ‘beware, think’ mentioned above and related
with the aforementioned verb maudžiù, maũsti ‘ache, long for’.

6 PSl. *naglъ

PSl. *naglъ̀ (a.p. b) ‘sudden, impetuous, impudent’ is reflected as OCS naglo

(adv.) ‘very quickly’, Slov. nágel ‘hasty’, Croat. nȃgao ‘hasty’, Russ. náglyj ‘cheeky,
impudent’, Pol. nagły ‘sudden, urgent’, Cz. náhlý ‘sudden, hasty’, etc. (Derksen
2008: 345, ESSJa 22: 33). This word looks like a verbal adjective of some verbal
root, to which the prefix *na- ‘on’ had been added. It has been connected with
the root of the adjective *golъ ‘naked’ (OCS golъ, Russ. gólyj, Croat. gȏl, etc.),
and the semantic development would have been from ‘naked’ to ‘insolent, im-
pudent’, which would imply that the oldest meaning is preserved in Russian
(cf. also Bulg. náglo ‘insolently’). Like the connection with *nagъ ‘naked’ (OCS
nagъ, Russ. nagój, Croat. nȃg, etc.), this is semantically difficult, and formally
without parallel (the prefix *na- would have been added to the zero-grade of
the adjectival root *gl-). The connection with Skr. áñjas ‘quickly’, Goth. anaks
‘id.’ (Skok 1985: 498, Vasmer 1953–1958, 2: 193) is improbable both formally (one
would need to assume Schwebeablaut *Heng-/*Hneg-) and semantically, as the
Russian meaning ‘cheeky, impudent’ is difficult to derive from the original
‘quick’. Therefore, I would like to propose a new etymology.
It is possible to regularly derive PSl. *naglъ̀ from the negated adjective *ne-
-h₂egh-lo- ‘who is not afraid’, from the PIE root *h₂egh- ‘be afraid, be sad’ (Goth.
un-agands ‘fearless’, OIr. -ágor ‘I am afraid’). The semantic development was
from ‘who is not afraid, fearless’ to ‘haughty’ and ‘insolent, cheeky’, which is
the original meaning in Slavic, still preserved in Russian. For the form of the
negative prefix *ne- cf. below on PSl. *němъ.
7 PSl. *němъ 111
Matasović
PSl. *němъ (a.p. c) ‘mute, dumb’ is reconstructed on the basis of OCS němъ, Cro-
at. nijȇm, Russ. nemój, Ukr. nimýj, Slov. nȇm, Cz. němý, Pol. niemy (cf. ESSJa 25:
100ff.). The etymology of this Slavic word is disputed. Some derive it by dissim-
ilation from *mēm- (cf. Latv. mḕms ‘dumb’, which may be a Slavic loanword), cf.
Snoj (2003: 442), Vasmer (1953–1958, 2: 212). Originally, the root *mVm- appears
to be onomatopoetic, cf. Lat. mūtus ‘dumb’, Germ. mummeln, and in Slavic it is
represented without the dissimilation by Croat. mumljati ‘mumble’. Although
it is not impossible, both the assumption of the dissimilation (*mēm- > *nēm-)
and the lengthened grade in this root are quite ad hoc. Likewise, the derivation
from *neh₁emo, the negated root *h₁em- ‘get, take’ (Lat. emō ‘buy’, Lith. imù, im̃ti,
OCS imati, etc., IEW 301f.) requires the assumption that there was a complex
semantic development (*‘who does not get (it), who does not understand’ >
*‘stupid’ > ‘dumb, who cannot speak’). For these reasons, I believe it is fair to
say that the etymology of PSl. *němъ is unclear.
However, I believe that there is an alternative etymology that does not have
to rely on irregular dissimilation, or speculative semantic development. PSl.
*němъ can be regularly derived from PIE *ne-h₁g’-mo-, the negated adjective
derived from the root *h₁eg’- ‘speak, say’ (Lat. aiō ‘say’, Arm. asem, Gr. ēmí, IEW
290–291). The phonological development would be regular, as *g is regularly
lost before *m in Slavic (Matasović 2010), cf. *teg-men- > PSl. *těmę ‘top of the
head, sinciput’ (Russ. témja, Croat. tjȅme, etc.), from the root *teg- ‘cover’ (Lat.
tegō, Gr. stégō, etc.). The semantic development from ‘not speaking, who is un-
able to speak’ to ‘dumb’ would have been straightforward. However, it must
be admitted that the problem for this etymology is the form of the negative
prefix *ne- instead of the more common zero-grade *n-, since *nh₁g’mo- would
probably have yielded *ęmъ in Proto-Slavic. However, the full grade *ne- does
occur in Slavic nominal word-formation, cf. the adjective *nebogъ ‘poor’ (OCS
nebogъ) from *ne- and the nominal root *bogъ ‘god’ (< *‘fortune’).

References

Beekes 2010: Beekes, Robert S. P., Etymological Dictionary of Greek, Leiden: Brill.
Derksen 2008: Derksen, Rick, Etymological Dictionary of the Slavic Inherited Lexicon, Leiden:
Brill.
De Vaan 2008: De Vaan, Michiel, Etymological Dictionary of Latin and the other Italic Languages,
Leiden: Brill.
ESSJa: Etimologičeskij slovar’ slavjanskix jazykov, 1–39–, Moskva: Nauka 1974–2014–.
Gluhak 1992: Gluhak, Alemko, Hrvatski etimološki rječnik, Zagreb: August Cesarec.
IEW: Pokorny, Julius, Indogermanisches etymologisches Wörterbuch, Bern: Francke 1959.
 112 Illič-Svityč 1963: Illič-Svityč, Vladislav M., Imennaja akcentuacija v baltijskom i slavjanskom,
Matasović Moskva: Izdatel’stvo AN SSSR.
LIV: Rix, Helmut et alii, Lexikon der indogermanischen Verben, Wiesbaden: Reichert 2001.
Matasović 1998: Matasović, Ranko, The Proto-Indo-European Heteroclita in Balto-Slavic.
Wiener Slavistisches Jahrbuch 44, 121–127.
Matasović 2009: Matasović, Ranko, Etymological Dictionary of Proto-Celtic, Leiden: Brill.
Matasović 2010: Matasović, Ranko, The Loss of *g before *m in Proto-Slavic. Rasprave Instituta
za hrvatski jezik i jezikoslovlje 36/2, 299–306.
Matasović 2013: Matasović, Ranko, Substratum words in Balto-Slavic. Filologija 60, 75–101.
Matasović 2014: Matasović, Ranko, Slavic Nominal Word-Formation, Heidelberg: Winter.
REW: Meyer-Lübke, Wilhelm, Romanisches etymologisches Wörterbuch, Heidelberg: Winter
1972.
Skok 1985: Skok, Petar, Etimologijski rječnik hrvatskoga ili srpskoga jezika, 2, Zagreb: HAZU.
Snoj 2003: Snoj, Marko, Slovenski etimološki slovar, Ljubljana: Modrijan.
Smoczyński 2007: Smoczyński, Wojciech, Słownik etymologiczny języka litewskiego, Vilnius:
Vilniaus Universitetas.
Varbot 1972: Varbot, Žanna Ž., Zametki po slavjanskoj etimologii. Etimologija 1970, 55–64.
Vasmer 1953–1958: Vasmer, Max, Russisches etymologisches Wörterbuch, 1–3, Heidelberg: Winter.
Young 2006: Young, Stephen, Slavic *gъrdъ : Lith. (pa)gìrtas. Baltistica 41, 371–374.