DOI 10.57589/srl.v70i2.4029
UDK 811.1/.2:81.37
Krzysztof Tomasz Witczak
Uniwersytet Łódzki (Univerza v Lodžu)
krzysztof.witczak@uni.lodz.pl
Grażyna Habrajska
Uniwersytet Łódzki (Univerza v Lodžu)
grazyna.habrajska@uni.lodz.pl
Mikołaj Rychło 
Uniwersytet Gdański (Univerza v Gdansku)
mikolaj.rychlo@ug.edu.pl
PROTO-SLAVIC COLLECTIVES IN *-ЬJE
FROM A HISTORICAL AND COMPARATIVE PERSPECTIVE1
The aim of this article is to conduct a diachronic analysis of Proto-Slavic collective nouns 
ending in *-ьje with a view to gathering comparative evidence for postulating its Indo-European 
origin. The formation at issue exhibits exact counterparts not only in Germanic, but also in 
Indo-Aryan and Tocharian—e.g., PSl. *berzьje n. coll. ‘birch woods, birch forest’ (= ON. birki 
n. ‘birch forest’ <PG. *berkijan n. coll.), PSl. *bratrьje n. coll. ‘brothers, brotherhood’ (= Ved. 
bhrātryám n. coll. ‘brotherhood’, Toch. A pratri ‘brothers’), PSl. *govьje n. coll. ‘horned cattle’ 
(= Ved. gávyam, gavyám n. coll. ‘herd of cows’); PSl. *vьrmьje n. coll. ‘worms’ (= ON. yrmi n. 
coll. ‘id.’ < PG. *wurmijan n. coll. ‘swarm of worms’). The authors polemicize against Ranko 
Matasović (2005), who assumes a secondary origin of the collective *-ьje, and argues that the 
suffix stems from the Indo-European collective nouns in *-i- (whence PSl. *-ь).
Keywords: archaism, collectives, etymology, Germanic-Balto-Slavic relations, Proto-
Indo-European, Slavic languages
Namen članka je zgodovinsko primerjalna analiza praslovanskih kolektivnih (zbirnih) 
samostalnikov na *-ьje z dokumentacijo indoevropskega primerjalnega gradiva, ki bi kazalo na 
indoevropski izvor teh tvorjenk. Slednje imajo namreč natančne vzporednice ne le v german-
ščini, temveč tudi v stari indijščini in toharščini, npr. psl. *berzьje s kol. 'brezov gozd, brezje' 
(= stnord. birki n. 'brezov gozd, brezje' < pgerm. *berkijan s kol.), psl. *bratrьje s kol. 'bratje, 
bratovščina' (= ved. bhrātryám s kol. 'bratovščina', toh. A pratri 'bratje'), psl. *govьje s kol. 'rogata 
živina' (= ved. gávyam, gavyám s kol. 'čreda krav'); psl. *vьrmьje s kol. 'črvad' (= stnord. yrmi 
s kol. 'isto' < pgerm. *wurmijan s kol. 'množica črvov'). Avtorja polemizirata z interpretacijo 
Ranka Matasovića iz leta 2005, ki predpostavlja drugotni izvor kolektivne (zbirne) pripone 
*-ьje, domnevno nastale pod vplivom indoevropskih kolektivov na *-i- (od koder je psl. *-ь).
1 The present article is part of a research project entitled Collective nouns in the Slavic and Germanic 
languages, financed by the scholarly development fund of the Faculty of Philology, University of Lodz. We 
are indebted to Penny Shefton for proofreading the whole text and making several stylistic improvements, 
Silvo Torkar for translating the English abstract into Slovenian, as well as to anonymous reviewers for useful 
remarks. The full responsibility for all opinions and errors that still remain lies solely with us. 
Slavistična revija, letnik 70/2022, št. 2, april–junij158
Ključne besede: arhaizem, kolektivni (zbirni) samostalniki, etimologija, germansko-balt-
sko-slovanski odnosi, praindoevropščina, slovanski jeziki
1 Introduction
The aim of this paper is to argue that Slavic collective nouns in *-ьje are much 
older than assumed so far. Researchers that have investigated this collective formation 
(e.g. Vaillant 1964: 208; Sławski 1974: 85–86; Matasović 2005: 35–38) conclude that 
it arose in Proto-Slavic, though they agree that the Proto-Slavic suffix *-ьje and the 
abstract nouns in *-ьje are probably of Indo-European origin2. The present paper at-
tempts to disprove some of their assumptions and to present new evidence supporting 
the Indo-European origin of the collectives in *-ьje, suggested originally in two earlier 
works (Habrajska, Witczak 1994: 31–39; Habrajska, Rychło, Witczak 2020: 261–83). 
After a brief introduction, the findings of the previous research are reported: Franciszek 
Sławski’s (1974) in the second section and Ranko Matasović’s (2005) in the next one. 
The fourth section presents selected comparative material coming from Indo-Aryan 
languages, Hellenic and Tocharian. The fifth section offers additional evidence by jux-
taposing the Slavic collective nouns at issue with arboreal collectives attested mainly 
in North Germanic languages. The sixth section demonstrates further evidence coming 
from East and West Germanic. The seventh section discusses a possible relationship 
between Slavic collectives in *-ьje and those in *-ь. In the eighth section, we postu-
late an Indo-European origin of the Slavic collective nouns in *-ьje by outlining their 
development with four phases of derivation. The final section presents the conclusion. 
From a descriptive point of view, collective nouns (nomina collectiva) are de-
nominal derivatives with a singular form and a collective function revealing the sense 
‘multiplicity, collectivity of what is specified in the base’ (Habrajska 1995: 313–16), 
e.g. Pol. brzezie n. ‘a group of birches; a birch forest’ ← Pol. brzoza ‘birch (a decidu-
ous tree), Betula L.’. Within collective nouns, the derivational type in *-ьje deserves 
closer attention, as Borek (1988: 94) put it, “because of its pre-Slavic past as well as 
its chronologically diverse vitality” (in Polish; our translation). The evidence coming 
from the oldest Slavic manuscripts allows us to believe that this derivational type gained 
special productivity in the lexicon related to the plant world (Bernstein 1985: 362). 
The Old Church Slavonic examples include: bylije ‘grass’, vrьbije ‘willow’, dǫbije 
‘trees’ etc. Likewise, in Old Russian texts, one can find numerous plant collectives e.g. 
ORu. derevije ‘trees’, kropivije ‘nettle’, kustovije ‘bushes, shrub thickets’, smokvije ‘fig 
trees’. Undoubtedly, one should agree with Franciszek Sławski (1974: I 86; 2011: 61), 
who recognized the suffix *-ьje, as “the most productive derivational affix forming 
collectives primarily from the names of inanimate objects, especially from the names 
of trees, shrubs, berries” (in Polish; our translation).
2 Cf. Sławski (1974: 86): “Przyrostek -ьje jest dziedzictwem ie., kontynuuje ie. *-ii̯̯o-” [The suffix -ьje 
is an Indo-European heritage; it continues IE. *-ii̯o-]; Matasović (2005: 37): “The suffix [...] *-ьje was 
presumably inherited in some words, and then spread analogically to others, thus becoming productive”.
159Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
Collectives in *-ьje denoting a group of trees appear frequently in the Serbo-Croatian 
lexicon (Frančić 1961: 36–40) and toponymy (Mianowicz 1976: 55–66), as well as in 
the Slovenian language (Bajec 1950: 117); they are less extensively attested in Czech 
and Slovak, as well as in the Lachian dialects (Laskowski 1996: 71–73). 
In other Slavic languages, among others in Polish, the collectives in *-ьje have 
almost completely lost their vitality, becoming an unproductive formation. In relation 
to the Polish language, it can be observed that this old derivational type of collectives 
in the language of the 16th-17th centuries becomes less productive, especially in the 
case of collectives denoting a group of trees (Rykiel-Kempf 1985: 21). According to 
Stanisław Rospond (1971: 197), what contributed to the decline in use of Old Polish 
singular collective names was their partial identification with plural forms of the type 
as exemplified in Pol. węgle m. pl. ‘coals’ (from węgiel ‘coal’) and liście m. pl. ‘leaves’ 
(from liść ‘leaf’). It should be noted, however, that the traces of collective nouns derived 
from the names of trees have been preserved in a rich repertoire of Polish toponymy 
(Bańkowski 1972: 281–86; Borek 1988: 87–96; Kreja 1988: 41–82; Różycka 1989: 
53–99; Lech-Kirstein 2015: 223–39). 
2 Franciszek Sławski (1974) on the origin of Slavic collectives in *-ьje 
The origin of Slavic collective nouns in *-ьje remains without explanation. It is usually 
argued that the derivational type in *-ьje is not uniform either in origin or in semantics. 
Furthermore, apart from (a) singular collectives in *-ьje, we can also distinguish: (b) 
deadjectival abstract nouns of the type *obilьje ‘abundance’ (: *obilъ ‘abundant’) and 
(c) formations from prepositional expressions that perform a relational-localizing role, 
e.g. Pol. poddasze ‘room under the roof’ (← Pol. dach ‘roof’), zarzecze ‘area behind 
the river’ (← Pol. rzeka ‘river’), zawodzie ‘area behind the water’ (← Pol. woda ‘wa-
ter’) (Borek 1988: 94). There are a few Proto-Slavic abstract nouns derived from other 
nouns, e.g. PSl. *sъnьje n. ‘dream’ ← *sъnъ m. ‘sleep’; PSl. *ustьje n. ‘mouth of the 
river’ ← *usta n. pl. ‘mouth’; PSl. *znamenьje n. ‘sign’ ← *znamę n. ‘id.’. 
Franciszek Sławski traces the collective function of the suffix *-ьje to the primary 
adjectival abstract formations of the type *obilьje ‘abundance’, which, when they became 
concrete, gave rise to later collectives derived from the names of inanimate objects, 
especially the names of trees and shrubs. Moreover, when considering the suffix as a 
case of Indo-European heritage, Sławski derives PSl. *-ьje from IE. *-ii̯o-m, and also 
supposes that it could originally be a substantivized neutrum of adjectives in *-ii̯o- 
(Sławski 1974: I 86; 2011: 61). All in all, the development of the Slavic collectives in 
*-ьje, according to Sławski, looks as follows: (A) a specific nominal formation ⇒ (B) 
denominal adjective formed from the noun in question ⇒ (C) adjectival abstractum in 
*-ьje ⇒ (D) collective noun in *-ьje. 
Although Sławski did not formulate his view explicitly, it seems that in his approach, 
the singular collective nouns in *-ьje represent a typically Slavic formation which 
probably evolved in the early Proto-Slavic era. 
Slavistična revija, letnik 70/2022, št. 2, april–junij160
3 Matasović on Slavic collectives in *-ьje 
The Croatian linguist Ranko Matasović (2005) stresses (like Sławski) that the 
Proto-Slavic nominal suffix *-ьje is commonly used for deriving neuter nouns with 
abstract or collective meaning. The basic difference consists in the process of their 
derivation. According to Matasović (2005), the Proto-Slavic abstract nouns derive from 
adjectives (e.g. PSl. *veselьje n. abstr. ‘joy’ ← *veselъ adj. ‘merry, gay’), whereas 
the Proto-Slavic collectives are formed from other nouns (e.g. PSl. *kamenьje n. coll. 
‘a heap of stones’ ← *kamy, gen. sg. *kamene m. ‘stone’). Unfortunately, he ignores 
the fact that numerous abstract nouns, attested in the Slavic and other Indo-European 
languages, derive not only from adjectives, but also from simple nouns. It is worth 
emphasizing that most Slavists and Indo-Europeanists suggest the Indo-European 
origin of the suffix *-ьje on the basis of the following correspondences: PSl. *ustьje 
n. ‘mouth of the river’ = Lat. ōstium n. ‘id.’; PSl. *sъnьje n. ‘dream’ = Lat. somnium 
n. ‘id.’, Gk. ἐνύπνιον n. ‘vision (during sleep); dream’ (Brugmann, Delbrück 1906: 
187–89; Meillet, Vaillant 1934: 357–58; Sławski 1974: I 86; 2011: 61; Jurišić 1992: 
52–4). The two comparisons (which do not represent collective nouns but derive from 
nouns, i.e. PSl. *usta and PSl. *sъnъ, respectively) seem to demonstrate the same neuter 
suffix *-ii̯om (Habrajska, Witczak 1994: 32–8), which could be treated as a variant of 
the adjectival morpheme *-ii̯o-. 
According to Ranko Matasović, the “putative” suffix *-ii̯o- (or *-iyo- in Matasović’s 
notation) should be rejected for two reasons. First, “there was never a suffix *-iyo- in 
PIE. [...] Only in Old Indic do we find the disyllabic form -iya-; however, this form of 
the suffix is regular in nouns derived from the feminine stems in -ī- < *-iH- (Burrow 
1973: 186), but it was extended analogically to other stems as well. Thus, OInd. -iya- is 
actually derivable from *-iHo-” (Matasović 2005: 35–63).
This argument can be valid, if we accept that the feminine gender existed in the 
oldest phase of Proto-Indo-European. The feminine gender is not commonly recon-
structed in the Indo-European linguistics. Some historical linguists believe that the 
Anatolian languages preserve the original distinction of two genders only, namely the 
commune (or animate) gender and the neuter (or inanimate) one. The adjectival suffix 
*-ii̯o- is clearly attested in most Indo-European languages (Krahe, Meid 1967: 71; Fritz, 
Meier-Brügger 2021: 284). Of course, it is observed that the Old Indic forms frequ-
ently demonstrate the analyzed suffix in two cognate forms -ya- and -iya-. However, 
the simplified (monosyllabic) version -ya- clearly derives from an earlier disyllabic 
archetype -iya-. This phenomenon cannot be treated as an argument for deriving of the 
suffix *-ii̯o- from the feminine suffix *-ih2. 
Matasović’s argument devoted to the “putative” suffix *-ii̯o- seems doubtful for 
two reasons. Firstly, there are no obvious Proto-Indo-European adjectives in *-ii̯o- or 
*-ih2o- derived from feminine ih2-stem nouns. For example, Greek πότνια f. ‘mistress, 
lady’ (< PIE. *pótnih2 f. ‘id.’, cf. Ved. pátnī f. ‘female possessor, mistress; wife’) 
3 There is an added page (p. 186) in the quotation. 
161Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
demonstrates only the feminine adjective belonging to the d-stem (cf. Gk. ποτνιάς, 
-άδος adj. ‘venerable, august, powerful’). Moreover, no Indo-Aryan adjective derived 
directly from pátnī is attested. 
Secondly, Proto-Indo-European adjectives in *-ii̯o- frequently derive from nouns 
of the masculine or neuter gender, cf. 
3.1 Gk. ἄγριος adj. ‘of the fields, wild, savage (of animals), uncultivated (of lands); 
savage, violent, cruel (of persons)’, Ved. ajryàḥ adj. ‘being in or connected with a field 
or plain’ < PIE. *h2eǵrii̯ós adj. ‘id.’ ← PIE. *h2eǵros m. ‘field’, cf. Lat. ager m. ‘field’, 
Gk. ἀγρός m. ‘field, land, farm’, Ved. ájra- m. ‘field, plain’. 
3.2 Ved. divyáḥ adj. ‘divine, heavenly, celestial’, Gk. δῖος adj. ‘celestial, divine, 
noble, magnificent; of Zeus’, Myc. Gk. di-wi-jo ‘(in the month) of Zeus’ (Burrow 1973: 
185; Fritz, Meier-Brügger 2021: 285) < PIE. *diu̯ii̯ós adj. ‘celestial; belonging to the 
Sky-god’ ← PIE. *diēu̯-s m. ‘Sky-god’,
3.3 Gk. πάτριος adj. ‘of the father; paternal, traditional’, Lat. patrĭus adj. ‘belonging 
to the father; paternal’, Ved. pítryaḥ adj. ‘derived from a father; relating to a father; 
paternal, patrimonial, ancestral’ (Burrow 1973: 185; Fritz, Meier-Brügger 2021: 285) < 
IE. *pə́2trii̯os adj. ‘paternal’ < PIE. *ph̥2trii̯os adj. ‘paternal’ ← PIE. *ph̥2ter-s m. ‘father’. 
3.4 Ved. rājyáḥ adj. ‘kingly, princely, royal’, Lat. rēgĭus adj. ‘belonging to a king; 
kingly, royal, regal’ (Burrow 1973: 185) < PIE. *reh1ǵii̯ós adj. ‘of a king; royal’ ← 
PIE. *reh1ǵ-s m. ‘king’. 
Of course, we can find adjectives derived from nouns being categorized as either 
feminine or the common gender (i.e. both the feminine and the masculine). It is worth 
emphasizing that all the Anatolian languages preserved the commune, i.e. the animate 
gender.  
3.5 Gk. χθόνιος, Ved. kṣámyaḥ adj. (Burrow 1973: 185) < PIE. *dhǵhómii̯os adj. ← 
PIE. *dhǵhóm-s f. ‘earth’, cf. Gk. χθών f. ‘earth’; Ved. kṣā-́ f. ‘id.’; Hitt. tekan- n. (gen. sg. 
taknaš) ‘earth’, Luw. tiia̯mmiš n. ‘id.’, HLuw. takam- ‘id.’ (Kloekhorst 2008: 858–59). 
3.6 Ved. gávyaḥ adj. ‘consisting of cattle or cows, coming from or belonging to 
a cow’ (Burrow 1973: 185; Monier-Williams 1999: 351) < IE. *gu̯óu̯ii̯os adj. ← PIE. 
*gu̯h3éu̯-s f./m. ‘cow; ox’, cf. Ved. gáuḥ f./m. ‘id.’; Gk. βοῦς f./m. ‘id.’, Luw. u̯au̯a- c. 
‘cow’; HLuw. wawa- c. ‘id.’, Lyc. wawã acc. sg. ‘cow’ (Kloekhorst 2008: 507). 
It is worth emphasizing that the abstract nouns in *-ii̯om do not seem to derive from 
the Proto-Indo-European feminine ih2-stems. For example, Lat. somnium n. ‘dream’ 
derives from Lat. somnus m. ‘sleep’ (< PIE. *su̯ópnos m. ‘id.’), whereas PSl. *sъnьje 
n. ‘dream’ and Gk. ἐνύπνιον n. ‘vision (during sleep); dream’ go back to the common 
archetype *súpnos m. ‘sleep’, cf. PSl. *sъnъ m. ‘id.’, Gk. ὕπνος m. ‘id.’. 
Slavistična revija, letnik 70/2022, št. 2, april–junij162
Our conclusion is that the adjectival suffix *-ii̯o-, attested in most Indo-European 
languages, has nothing to do with the feminine ih2-stem nouns. In other words, there is 
no convincing reason for rejecting the neuter suffix *-ii̯om, which creates both abstract 
and collective formations. 
4 The Slavic collective nouns in *-ьje and -ьja 
In his paper Matasović introduces two arguments for the secondary origin of the 
Slavic collectives in *-ьje, assuming that “old PIE neuters with barytone accentuation 
became masculines in Proto-Slavic, as Illič-Svityč has shown (1963), e.g. *dhwórom 
»gate« (OInd. dvā́́ram, L forum) > OCS dvorъ. Therefore, all of the putative cognates 
of Slavic derivatives in -ьje would have to be oxytona, and in Old Indic we find that 
neuters in -iya- are, as a rule, never oxytona (Burrow 1973: 185)” (Matasović 2005: 36). 
He refers to the Indo-Aryan lexical evidence, as well as the accentual argument. 
It is necessary to examine the correctness of Matasović’s words. There is a valuable 
pair of the related Slavic collectives in *-ьja and -ьje, demonstrating exact cognates 
in other Indo-European languages:
4.1 PSl. *bratrьja f. coll. ‘brothers; brotherhood’ (cf. OCS. братрия, OSln. bratria, 
OČak. (15th c.) bratrja, ORu. братрия f. coll., OPol. bratrza f. coll., LSorb. (obso-
lete) bratśa, USorb. bratřa pl. ‘brothers’, Cz. bratří pl. ‘brothers’, Slovak braträ pl. 
‘brothers’) = Lith. dial. (in Kurschat’s dictionary) brotìja, brotijà gen. sg. brotìjos f. 
‘circle of the most faithful friends’ (< PBalt. *brātríjā f. coll. ‘brotherhood, brothers’ 
by means of the progressive dissimilation of the liquid stops *r–r > *r–ø)4; Gk. Att. 
φρᾱτρίᾱ f. coll. ‘family group; clan, phratry; a subdivision of the phyle in Athens’, dial. 
φᾱτρίᾱ f. coll. ‘id.’ (Liddell, Scott 1996: 1953; Montanari 2018: 2307; Diggle 2021: 
1477). The above-mentioned Slavic, Baltic and Greek forms go back to the collective 
noun *bhreh2tríi̯eh2 f. coll. ‘a group of brothers; brotherhood’ (← PIE. *b
hréh2ter-s m. 
‘brother’) (Pokorny 1959: 164; Derksen 2008: 60; Matasović 2014: 145). The Greek 
and Balto-Slavic accentual evidence seems to indicate the barytone or the paroxytone 
stress of primitive collectives of the feminine gender. 
4.2 PSl. *bratrьje n. coll. ‘brothers; brotherhood’ (cf. Ru. dial. братьё -тья coll., 
Bel. брáтрия n., Ukr. брáття n. ‘id.’) = Ved. bhrātryám n. coll. ‘brotherhood, fra-
ternity’. The East Slavic and Indo-Aryan forms derive from PIE. *bhreh2trii̯óm n. coll. 
‘a group of brothers; brotherhood’ (← PIE. *bhréh2ter-s m. ‘brother’). See also Ved. 
bhrātrám n. coll. ‘brotherhood, fraternity’ (Monier-Williams 1999: 770). The Slavic 
and Vedic accentual evidence seems to document the original oxytone stress in the 
neuter collective noun in question.
4 There is an analogous collective noun brolijà, gen. sg. brolìjos f. ‘brothers and sisters, siblings; 
brotherhood, society’ in the literary Lithuanian language (Smoczyński 2018: 131). It derives from the di-
minutive noun brólis, gen. sg. brólio m. ‘brother’ (originally a shortened form of Lith. broterė̃lis m. ‘little 
brother’, cf. Žem. brótis m. ‘brother’, OPrus. brote ‘brother’).
163Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
Moreover, Toch. A pratri ‘brothers’ has to represent a collective noun, derived from 
the East Tocharian noun pracar m. ‘brother’, cf. Toch. B procer m. ‘id.’ (Witczak 2016: 
126–30). It is uncertain, however, whether Toch. A pratri ‘brothers’ should be derived 
from the former Proto-Indo-European archetype (4.1) or the latter (4.2). 
The Vedic language demonstrates also two further collectives in *-(i)yā (f. coll.) 
and *-(i)yam (n. coll.), related to each other:
4.3 Ved. (only in Pāṇini) gávyā f. coll. ‘cow-herd’ = Lith. gaujà f. coll. ‘flock, pack, 
herd, bunch, band, gang’; Latv. gauja f. ‘crowd’, Gk. Lac. βοῦα f. coll. ‘a band of young 
Spartan boys’ (as if from PGk. *βόϝιι̯ᾱ) (Kaczyńska 2019). These forms derive from 
the collective noun (IE.) *gu̯óu̯ii̯eh2 f. ‘a group of cows’ (← IE. *g
u̯óu̯ii̯os adj. ← PIE. 
*guh̯3éu̯-s f./m. ‘cow; ox’). 
4.4 Ved. gávyam, also gavyám n. coll. ‘herd of cows’, attested six times in hymns of 
RigVeda, e.g. RV I 140.13; V 34.8; VII 18.7; IX 62.23 (Monier-Williams 1999: 351), 
shows also a number of reflexes in modern Indo-Aryan languages, e.g. Oriya gāba 
‘cattle’, also ‘a cow’ as the effect of reducing the old collective form to the singular 
(Turner 1966: 219). The Indo-Aryan collective noun in question points undoubtedly to 
an original oxytonesis. The barytone variant gávyam seems to demonstrate a secondary 
accent taken from the adjective gávyaḥ adj. ‘consisting of cattle or cows, coming from 
or belonging to a cow’ (3.6) or from the feminine collective gávyā ‘cow-herd’ (4.3). 
The exact equivalent of the Vedic collective noun is attested in Proto-Slavic *govьje 
n. coll. ‘cattle, pecus’, attested exclusively in BRus. dial. гóўе, гаўя́ n. coll. ‘horned 
cattle, cows; a generic term for cattle’ (Борысь 1979: 47–48; Boryś 2007; 275–76; 
Sławski 2001: VIII 158).
Three Vedic collective nouns of the neuter gender, namely bhratryám, bhratrám 
(4.2) and gavyám (4.4), clearly demonstrate an original oxytone stress. It is obvious, 
that the same primitive accent should be suggested for two Proto-Slavic collectives 
*bratrьje (4.2) and *govьje (4.4). This is why we have to conclude that Matasović’s 
words, quoted at the beginning of the fourth section, are misguided.
Moreover, the original oxytonesis also has to be reconstructed in some Germanic 
collectives in -ijan, e.g.
4.5 PG. *imbijan n. coll. ‘swarm of bees’ is confirmed by the following lexical items: 
OE. imbe n. ‘swarm of bees’; MDu. imme f. ‘bee’, OHG. impi m. ‘swarm of bees’, also 
‘bee’, MHG. imbe, imme m. ‘swarm of bees’. Guus Kroonen (2013: 117) suggests that 
“[t]he neuter OE. imbe continues a collective *imbja-, but OHG. impi and MDu. imme 
in the sense of ‘bee’ probably represent different formations, i.e. *imbja- and *imbjō”. 
The Proto-Germanic collective noun *imbijan (neuter o-stem) ‘swarm of bees’ seems 
to be related to Greek ἐμπίς (gen. sg. ἐμπίδος) f. (secondary d-stem) ‘gnat’ (Pokorny 
1959: 311). The primitive oxytone stress of PG. *imbijan (earlier *imfiján) documents 
Slavistična revija, letnik 70/2022, št. 2, april–junij164
the Proto-Indo-European archetype *h1empii̯óm n. coll. ‘swarm of insects (esp. gnats 
or bees)’. The place of accent is clearly indicated by Verner’s law.
For further examples of the original oxytonesis in Germanic collective nouns in 
*-ijan, see the examples quoted below: 5.3 (PG. *alizijan n. coll. ‘alder grove’, as if 
from *alisiján), 5.5 (PG. *lindijan n. coll. ‘lime forest, lime wood’ < *linþiján), 6.2 (PG. 
*fergunijan n. coll. ‘the range of mountains’ < *ferhwuniján). These Proto-Germanic 
formations unanimously demonstrate Verner’s effect caused by the primitive stress 
inherited from the Indo-European protolanguage. It cannot be questioned that the 
Indo-European stress was placed in the final (oxytone) position at least in these four 
Proto-Germanic collectives. In other words, Matasović’s accentual argument seems 
hardly conclusive.
5 Collective nouns in Germanic languages
Ranko Matasović introduces an additional argument for the alleged secondary origin 
of the Slavic collectives in *-ьje.
“Thirdly, the suffix *-(i)yo- is not used to form collectives in any other Indo-European 
language. If OCS sъnьje is indeed derivable from the same proto-form as L somnium, 
the specialization of collective meaning in Slavic would have to be accounted for.” 
(Matasović 2005: 36). 
Unfortunately, the above argument, so strongly pointed out in Matasović’s paper, 
is completely incorrect as well. As it was demonstrated above, there are at least two 
collective nouns in *-ii̯om in Indo-Aryan, which were mentioned in the preceding secti-
on (see 4.2: Ved. bhrātryám n. coll. = PSl. *bratrьje n. coll. ‘brothers; brotherhood’; 
4.4: Ved. gavyám n. coll. = PSl. *govьje n. coll. ‘cows; herd of cows’). Moreover, we 
are able to quote numerous collective formations in *-ijan in the Germanic languages 
(see 4.5 and examples quoted in Sections 5 and 6). Some of them must be treated as 
exact equivalents of the Slavic collectives (Habrajska, Rychło, Witczak 2020: 271–74). 
Below we quote six Germanic collective formations (5.1–6) having exact or relatively 
close Slavic cognates (see 5.1a–6a).
5.1 PG. *berkijan n. coll. ‘birch forest, birch grove’ is clearly reflected in the 
Nordic languages: ON. birki n. ‘birch forest’, Icel. birki ‘birch forest; birch’, Norw. 
birki ‘birch forest’, Sw. björke ‘birch forest, birch grove’ (Falk, Torp 1910: 74–75, 
s.v. Birk; Jóhannesson 1956: 623; de Vries 1977: 37). The tree collective undeniably 
derives from the Proto-Germanic arboreal term *berkō f. ‘birch’, cf. ON. bjǫrk f. ‘birch’, 
Norw. bjørk ‘id.’, Sw. björk ‘id.’; OE. beorc f., Du. berk c.; OHG. biricha, G. Birke f. 
‘id.’ (Kroonen 2013: 61), by means of the Proto-Germanic suffix *-ijan. 
5.1.A PSl. *berzьje n. coll. ‘birch grove, birch forest’ (Трубачев 1974: I 208; 
Sławski 1974: I 213) demonstrates an extensive attestation and a large distribution in 
the Slavic languages, cf. Ru. берéзье n. ‘birch forest, birch twigs’; OPol. brzezie n. 
165Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
‘birch forest’, Pol. brzezie n. ‘birch grove or forest’; Cz. březí n. ‘small birch-grove’, 
also břízí n. ‘birch twigs, birch-wood’; Sla. brezie n. ‘small birch-forest, birch-grove’; 
SC. brȇzje n. ‘birch forest’, Sln. brẹ̑zje n. ‘id.’ (Habrajska, Witczak 1994: 35).
The Proto-Slavic collective noun *berzьje ‘birch forest, birch grove’ is identical to 
PG. *berkijan n. coll. ‘id.’ from the point of view of the semantics and Indo-European 
word-formation. The Germanic and Slavic formations regularly go back to the Proto-
Indo-European archetype *bherh̥2ǵii̯om n. coll. ‘birch grove, birch forest’, originally ‘a 
group of birch trees’ (Habrajska, Rychło, Witczak 2020: 271, 278).
5.2 PG. *alizijan n. coll. ‘alder grove’ can easily be reconstructed on the basis of 
ON. elri n. ‘alder grove’ and Icel. elri n. ‘alder-tree’ (de Vries 1977: 101; Kroonen 
2013: 22). It derives from PG. *alisō / *alizō f. ‘alder’, cf. Sp. aliso m. ‘alder tree’ (a 
Gothic loanword); Du. els c., OE. alor, E. alder; OHG. elira, also erila, G. Erle ‘alder’ 
(Kroonen 2013: 22). 
5.2.A PSl. *olьšьje / *jelьšьje n. coll. ‘alder grove, alder forest’ is attested in East 
Slavic (cf. Ru. dial. олéшье n. ‘alder grove’, алéшье n. ‘alder-tree’; Bel. алéшша n. 
‘alder grove’), West Slavic (cf. OPol. olsze n. coll. ‘alder grove’; OCz. olšie n., Cz. olší 
n. ‘alder grove’, Sla. olšie n. ‘id.’; Rykiel-Kempf 1985: 21; Habrajska, Witczak 1994: 
34), as well as in South Slavic (cf. Sln. ǫ̑łšje n. ‘alder forest / Erlenwald’; SC. jȏšje n. 
coll., also jèlāšje, jelšje n. coll. ‘alder grove, alder forest’), cf. Pol. olsza f. ‘alder, Alnus 
L.’; Ru. ольха f. ‘id.’; SC. dial. jȅlša f. ‘id.’ (Habrajska, Witczak 1994: 34; Трубачев, 
Журавлев 2005: XXXII 83). 
It should be emphasized that the Proto-Slavic collective noun *olьšьje ‘alder gro-
ve, alder forest’ represents an exact equivalent of PG. *alizijan n. coll. ‘alder grove’ 
(Habrajska, Rychło, Witczak 2020: 272, 278).
5.3 ON. espi n. ‘aspen / Espe’, OSw. æspe- (attested in toponymy) and Sw. dial. 
äspe ‘Espenwäldchen / aspen grove’ represent a collective noun *aspijan denoting 
originally ‘aspen forest, aspen grove’ (Falk, Torp 1910: 25, s.v. Asp; de Vries 1977: 
106). It is evidently related to ON. ǫsp f. ‘aspen’, OE. æsp, E. aspen; Du. esp, OHG. 
aspa f. ‘aspen’, G. Espe f. ‘id.’ (< PG. *aspō, earlier *apsō f. ‘aspen, Populus tremula 
L.’) (Kroonen 2013: 39).
5.3.A PSl. *opsьje n. coll. ‘aspen forest, aspen grove’ (Трубачев, Журавлев 2005: 
XXXII 98) > Ru. dial. осьë n. ‘aspen forest, aspen grove / осиновый лес, осинник’, 
also ‘aspen-tree’ and ‘aspen-wood; aspen twigs’. The Proto-Slavic lexical item is also 
attested in Czech and Polish place names, cf. Cz. Vůsí, Osí, Vosí; Pol. Osie (Borek 
1988: 90; Bańkowski 2000: 434). The basic dendronym *opsa is reflected in USorb. 
wosa f. ‘aspen, Populus tremula L.’, LSorb. wósa f. ‘id.’; Pol. dial. osa f. ‘aspen’; Bel. 
асá f. ‘id.’5. 
5 There are many derivatives, especially diminutive formations in Proto-Slavic, e.g. PSl. *opsica f. ‘aspen’, 
PSl. *opsičina f. ‘id.’, PSl. *opsika f. ‘id.’ and PSl. *opsina f. ‘id.’ (Трубачев, Журавлев 2005: XXXII 93–6). 
Slavistična revija, letnik 70/2022, št. 2, april–junij166
PSl. *opsa f. ‘aspen’ resembles PG. *aspō f. ‘id.’ not only semantically, but also 
phonologically (Непокупный 1998: 35–43), though a metathesis of consonants (*ps > 
*sp) must be suggested for the Proto-Germanic language. By analogy, PG. *aspijan n. 
coll. ‘aspen forest, aspen grove’ may represent a metathesized variant of PSl. *opsьje 
n. coll. ‘id.’ (Habrajska, Rychło, Witczak 2020: 272, 278).
5.4 PG. *lindijan n. coll. ‘lime forest’, reflected exclusively in ON. lindi n. ‘Lindenwald, 
Lindenholz’ (Pokorny 1959: 677; de Vries 1977: 357), represents a collective noun 
derived from PG. *lindō f. ‘lime-tree, Tilia L.’, cf. ON. lind f. ‘lime-tree’, OSax. linda 
f. ‘lime-tree’, Du. linde c. ‘id.’; OHG. linta f., G. Linde f. ‘id.’ (Непокупный 1998: 
170–71). The Proto-Germanic dendronym goes back to the Indo-European archetype 
*lentā ́f. ‘a kind of tree or wood’ (< PIE. *lent-éh2), cf. Lith. lentà f. ‘board, plank’, Latv. 
lęñta f. ‘id.’ (Kroonen 2013: 338; Smoczyński, 2018: 688–89). The Proto-Germanic 
voiced stop *d (< IE. *t) clearly documents the effect of Verner’s law. In other words, 
both the basic term for ‘lime-tree’ and the related collective noun demonstrate an 
original oxytone stress (PG. *lindijan n. coll. < *linþiján n. coll. < PIE. *lentii̯óm). 
5.4.A PSl. *lǫtьje n. coll. ‘young lime grove’ (Трубачев 1990: XVI 162) > Ru. лутьë 
n. ‘young lime trees to be stripped of their bark’, dial. ‘young lime grove / молодый 
липняк’; Ukr. луття n. ‘young lime trees; twigs of a willow’; OCz. lútie n. ‘young 
lime tree; twigs’ (Фасмер 1986: 536; Habrajska. Witczak 1994: 35–6). 
It is obvious that the Proto-Slavic tree collective *lǫtьje ‘young lime grove’ cannot 
be separated from PG. *lindijan n. coll. ‘lime forest’. 
5.5 PG. *terwijan n. coll. ‘pine trees’ can be reconstructed on the basis of the 
following forms: ON. tyrvi n. ‘pinewood’, Norw. tyri ‘id.’, Sw. tyre, töre ‘dry wood, 
brushwood’. The Proto-Germanic tree collective in question is related to MHG. zirben 
m. ‘pine’ (< PG. *terwjan-). See also ON. tjara f. ‘tar’, OE. tierwe f. ‘tar’, E. tar; Du. 
teer n./c. ‘tar’ (Kroonen 2013: 514). 
5.5.A PSl. *dervьje n. coll. ‘growing trees; wood’ (Трубачев 1977: IV 213; Sławski 
1979: III 59–60) is clearly attested in most Slavic languages: ORu. деревиѥ n. ‘trees’; 
Ru. dial. дервьё́ n. ‘growing trees’, also ‘large, thick tree’; Bel. dial. дзерéвье n. ‘trees’; 
OUkr. деревье n. ‘id.’; OPol. drzewie n. coll. ‘growing trees’, also ‘wood’ (Habrajska 
1995: 139); OCz. dřievie, dřěvie n. ‘growing trees, forest; (fire)wood, a certain amount 
of wood’, Cz. dříví n. ‘(fire)wood, a certain amount of wood’, dial. ‘growing trees, 
forest’; Sla. drievie, dial. drevie n. (fire)wood OCS. дрѣвиѥ n. ‘trees, wood’; MBulg. 
дрѣвїе n. coll. ‘(growing) trees’; Sln. drẹ́vje n. coll. ‘trees / die Bäume’; SC. drȉjevlje, 
drȇvlje n. ‘growing trees; wood’ (Frančić 1961: 38). 
It can easily be argued that the Proto-Slavic collective noun *dervьje n. coll. 
‘growing trees; wood’ corresponds to PG. *terwijan n. coll. ‘pine trees; pinewood’ 
and goes back to IE. *deru̯ii̯om ‘a group of trees, esp. pine trees’ (Habrajska, Rychło, 
Witczak 2020: 271–72, 278).
167Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
5.6 ON. yrmi n. coll. ‘vermin’ < PG. *wurmijan n. coll. ‘swarm of worms or sna-
kes’ (Kroonen 2013: 600), cf. ON. ormr m. ‘snake’, G. Wurm ‘worm’, E. worm ‘id.’;6
5.6.A OCS. врьмиѥ n. coll. ‘insects’ and ORu. вéрмие n. coll. ‘vermin; insects 
/ Gewürm, Insekten’ (Фасмер 1986: 299) clearly derive from PSl. *vьrmьje n. coll. 
‘insects, vermin’.
According to Kroonen (2013: 600), the Old Norse collective noun yrmi “is formally 
identical” to the Old Church Slavic one. Theoretically, the Germanic and Slavic pair 
represents an Indo-European heritage, deriving from PIE. *u̯r̥mijom n. coll. ‘vermin; 
swarm of worms or insects’.
6 Other Proto-Germanic collectives 
Two further collectives in *-ijan are attested in the East and West Germanic lan-
guages. It may clearly be seen that they are not productive formations and they rarely 
refer to tree terminology. Generally, they seem to originate from the Proto-Germanic 
or perhaps even the Proto-Indo-European language. Some of them may illustrate traces 
of a Northern Indo-European vocabulary.
6.1 PG. *fanijan n. ‘fen, swamp’ is safely attested in most Germanic languages: Go. 
fani n. ‘mud’, ON. fen n. ‘fen, bog’, OE. fen n. (also m.) ‘marsh, mud’, E. fen; OSax. 
feni n. ‘fen’, MDu. veen, vene n. ‘bog’, Du. veen n. ‘bog’, ven n.’small lake’; OHG. 
fenni n. ‘swamp’, G. Fenn n. ‘id.’. The Proto-Germanic term in question represents, 
as Guus Kroonen (2013: 128) correctly stresses, a collective noun “potentially iden-
tical” to OPrus. pannean n. ‘ditch’ (< Proto-Baltic *panijan n.). The Germanic and 
West Baltic forms in question seem to go back to IE. *ponii̯om (Habrajska, Rychło, 
Witczak 2020: 268). 
6.2 PG. *fergunijan n. coll. ‘the range of mountains’ can be reconstructed on the 
basis of Go. fairguni n. ‘mountain range’ and OE. firgen n. ‘mountain’ (Kroonen 2013: 
136). It cannot be excluded that a related collective noun (of the feminine gender) 
additionally appears in OHG. Firgunnea f. ‘Ore Mountains’ (< PG. *ferguniō f. coll.). 
Note that a Proto-Germanic singulative seems to be attested in ON. fjǫrgyn f. ‘Mother 
Earth’ (< PG. *fergunī- f.).
7 The Slavic collectives in *-ьje and *-ь
Matasović (2005: 37) indicates that the Slavic languages demonstrate some neuter 
collectives in *-ьje which alternate with the feminine ones in *-ь or are related to an 
i-stem collective attested in other Indo-European languages, cf. Hitt. tuzziš c. i-stem 
‘people; army’ (< PIE. *teut-i-s); Latv. birzis f. i-stem ‘birch grove’, Lith. bìržė f. 
6 It is worth emphasizing that Germanists additionally reconstruct “Kollektivbildungen mit dem Prefix 
ga-” (Krahe, Meid 1967: 72), e.g. G. Gewürm n. ‘vermin’ < PG. *ga-wurmijan n. coll. ‘vermin; a swarm 
of worms’. 
Slavistična revija, letnik 70/2022, št. 2, april–junij168
coll. ‘birch wood, birch grove’ (← Lith. béržas m. ‘birch, Betula L.’, Latv. bȩ̃rzs m., 
bẽrze f. ‘id.’, OPrus. berse f. ‘birch’); PSl. *čędь f. coll. i-stem ‘offspring, children’ 
(← PSl. *čędo n. ‘descendant, child’) (Matasović 2014: 39). He finally concludes that 
most Slavic collectives in *-ьje (of the neuter gender) derive from the feminine i-stem 
nouns with the collective sense, though “[i]n many cases there is no evidence for an 
i-stem outside of Slavic” (Matasović 2014: 38). In other words, he agrees with his 
predecessors that the collective nouns in *-ьje attested in the Slavic languages represent 
a purely Proto-Slavic innovation. 
Matasović’s system of derivation of the Proto-Slavic collectives in *-ьje is supported 
by one example:
7.1 PSl. *ljudьje n. coll. ‘people’: OCS. ljudьje, Cr. ljȗdi, Pol. ludzie, now pl. ‘id.’ 
(Matasović 2014: 38). 
7.2 Lith. liáudis f. ‘rural people, simple people, peasants, a mob, a great crowd 
of people’, Latv. ļàudis f. coll. ‘people’; OE. lēod f. ‘id.’ (< PIE. *h1leud
his f. coll. 
‘people’). The Proto-Slavic evidence for an i-stem collective noun seems scarce and 
uncertain. Note that PSl. *ljudь f. ‘id.’ is attested only in Ru. dial. людь, -и f. coll. 
‘people, a lot of people’. 
Of course, the relatedness of the feminine i-stem nouns with the collective meaning 
to the neuter collectives in *-ii̯om is perfectly acceptable on the basis of Indo-European 
morphology. It is worth emphasizing, however, that such a derivation is connected 
with an unexpected change in the original gender. From the derivational point of view, 
it is possible to suggest that the original i-stem collective nouns (demonstrating the 
feminine gender in most Indo-European languages) should be easier to transform into 
the feminine collective nouns in *-ii̯eh2.
8 Indo-European derivation of the Slavic collectives
The lexical evidence presented in Section 5 and 7 clearly demonstrates that the 
Northern Germanic and Common Slavic collectives denoting ‘a group of trees’ or ‘a 
swarm of insects’ represent a purely Indo-European formation, the derivation of which 
can be simply explained from the point of view of the Indo-European word-formati-
on. Among various derivational functions of the Proto-Slavic suffix *-ьje, which are 
mentioned by Sławski (1974: I 86; 2011: 61), we would like to concentrate on the four 
which, we believe, can be arranged into the following phases of derivation:
 8.1 PIE. *bhréh2ters m. ‘brother’ (> IE. *b
hrā́tēr m. ‘id.’), e.g. Ved. bhrā́tā m. 
‘brother’; Lat. frāter m. ‘id.’; Go. brōþar m. ‘id.’; Gk. φράτηρ, Ion. φρήτηρ m. ‘member 
of the same phratry (i.e. a political and religious subdivision of the tribe)’; Pol. brat m. 
‘brother’ (< PSl. *bratrъ);
169Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
8.2 PIE. *bhréh2trii̯os adj. ‘pertaining to the brother’ (> IE. *b
hrā́trii̯os adj. ‘id.’), 
e.g. Gk. φράτριος, Ion. φρήτριος adj. ‘belonging to the phratry; protecting the phratry’; 
8.3 PIE. *bhréh2trii̯om n. abstr. ‘a quality or a state connected with the brothers, 
brotherhood, fraternity (in an abstract sense)’ (> IE. *bhrā́trii̯om n. ‘id.’), e.g. Gk. 
Φράτριον n. ‘temple used by a phratry’;
8.4 PIE. *bhreh2trii̯óm n. coll. ‘a group of brothers, brotherhood, fraternity’ (> IE. 
*bhrātrii̯óm n. ‘id.’): Ved. bhrātryám n. coll. ‘brotherhood, fraternity’; Ru. dial. братьё, 
братья coll. ‘brothers, brotherhood’, Bel. брáтрия n., Ukr. брáття n. ‘id.’ (< PSl. 
*bratrьje n. coll. ‘brothers, brotherhood’).
The derivational processes which must have been in play in this sequence can be 
reconstructed in the following way: starting from the final stage, PIE. *bhreh2trii̯óm, as 
a collective noun, is derived from the nomen abstractum *bhréh2trii̯om by a change of 
stress. It seems to be highly probable that the collective nouns in *-ii̯óm were derived 
from the related barytone abstracts in *-ii̯om (see 8.3) by an opposition of the original 
Indo-European stress.
This abstract noun, in turn, is derived from the adjective *bhréh2trii̯os by conversion 
(the stem remains intact, only the inflectional endings change). The earliest stage of 
derivation consisted in adjective formation, i.e. *bhréh2trii̯os from *b
hréh2ters.
An alternative way of creating collective nouns in Proto-Indo-European relies on 
the following feminine variant: 
8.5 PIE. *bhreh2tríi̯eh2 f. coll. ‘a group of brothers, brotherhood’ (> IE. *b
hrātríi̯ā f. 
coll. ‘id.’), e.g. Gk. Att. φρᾱτρίᾱ f. coll. ‘brotherhood; tribe, clan; a subdivision of the 
phyle in Athens’; Lith. dial. brotìja, brotijà f. ‘circle of the most faithful friends’; OCS. 
братрия, OSln. bratria, OČak. (XV w.) bratrja, ORu. братрия f. coll., OPol. bratrza 
f. coll., LSorb. (dated) bratśa, USorb. bratřa pl. ‘brothers’, Cz. bratří pl. ‘brothers’, 
Sla. braträ pl. ‘brothers’ (< BSl. *brātríjā f. coll. ‘brotherhood, brothers’). 
The feminine collectives in question seem to demonstrate the paroxytone stress 
falling on the short vowel -ĭ-, as documented by lexical data given above (8.5). In other 
words, the feminine collectives derived by means of the suffix *-íi̯eh2 were in accen-
tual opposition not only to the adjectives and abstract nouns in *-ii̯o- demonstrating 
barytone stress (see 8.2 & 8.3) but also to the neuter collectives in *-ii̯óm showing 
original oxytonesis (8.4). 
9 Conclusions 
Collective nouns in *-ьje, once productive in most Slavic languages, cannot be 
treated as a purely Proto-Slavic innovation. Like the feminine collectives in *-ьja, 
they seem to represent an Indo-European heritage. In fact, there are numerous exact 
Slavistična revija, letnik 70/2022, št. 2, april–junij170
or parallel formations in the Germanic and Indo-Aryan languages, as well as in the 
Hellenic and Tocharian subgroups.
Abbreviations
Languages & DiaLects
Att. – Attic; Bel. – Belorussian; BSl. – Balto-Slavic; Cz. – Czech; Du. – Dutch; E. 
– English; Elfd. – Elfdalian (Övdalian); Far. – Faroese; G. – German; Gk. – Greek; 
Go. – Gothic; Hitt. – Hittite; HLuw. – Hieroglyphic Luwian; Icel. – Icelandic; IE. 
– Indo-European; Ion. – Ionic; Lac. – Laconian; Lat. – Latin; Latv. – Latvian; Lith. – 
Lithuanian; LSorb. – Low Sorbian; Luw. – Luwian; Lyc. – Lycian; MBulg. – Middle 
Bulgarian; MDu. – Middle Dutch; MHG. – Middle High German; Norw. – Norwe-
gian; OČak. – Old Čakavian; OCS. – Old Church Slavic; OCz. – Old Czech; OE. 
– Old English; OHG. – Old High German; ON. – Old Norse; OPol. – Old Polish; 
OPrus. – Old Prussian; ORu. – Old Russian; OSax. – Old Saxonian; OSln. – Old 
Slovenian; OSw. – Old Swedish; PG. – Proto-Germanic; PIE. – Proto-Indo-European; 
Pol. – Polish; PSl. – Proto-Slavic; Ru. – Russian; SC. – Serbo-Croatian; Sla. – Slovak; 
Sln. – Slovenian; Sp. – Spanish; Sw. – Swedish; Ukr. – Ukrainian; USorb. – Upper 
Sorbian; Ved. – Vedic. 
gLosses
abstr. – abstract; adj. – adjective; c. – genus commune; coll. – collective; dial. – di-
alectal; e.g. – exempli gratia; f. – feminine; gen. – genitive; id. – idem / the same 
meaning; i.e. – id est / that is; m. – masculine; n. – neuter; pl. – plural; sg. – singular; 
s.v. – sub voce. 
references
Anton Bajec, 1950: Besedotvorje slovenskoga jezika. Vol. I. Ljubljana: SAZU. 
Andrzej Bańkowski, 1972: Rzeczowniki zbiorowe od nazw drzew w toponimii polskiej. 
Język Polski 52. 281–286. 
Andrzej Bańkowski, 2000: Słownik etymologiczny języka polskiego. Vol. II. Warszawa: 
Wydawnictwo Naukowe PWN. 
Samuil Borisovič Bernstein, 1985: Zarys gramatyki porównawczej języków słowiańskich. 
Alternacje. Tematy imienne. Warszawa: Państwowe Wydawnictwo Naukowe.
Henryk Borek, 1988: Górnośląskie nazwy miejscowe typu Bucze, Turze, Zawodzie. 
Wokół języka. Rozprawy i studia poświęcone pamięci profesora Mieczysława 
Szymczaka. Ed. Mieczysław Basaj. Wrocław, Warszawa, Kraków, Gdańsk, Łódź: 
Zakład Narodowy im. Ossolińskich. 87–96.
Wiesław Boryś, 2007: Etymologie słowiańskie i polskie. Wybór studiów z okazji 45-lecia 
pracy naukowej. Warszawa: Slawistyczny Ośrodek Wydawniczy.
171Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
Karl Brugmann, Berthold DelBrück, 1906: Grundriss der vergleichenden Grammatik 
der indogermanischen Sprachen. Vol. I. Teil. 1. Strassburg: Karl J. Trübner.
Thomas Burrow, 1973: The Sanskrit Language. London: Faber & Faber.
Rick Derksen, 2008: Etymological Dictionary of the Slavic Inherited Lexicon. Leiden, 
Boston: Brill.
James Diggle (ed.), 2021: The Cambridge Greek Lexicon. Vol. II. Cambridge: Cambridge 
University Press.
Hjalmar S. Falk, Alf torp, 1910–1911: Norwegisch-dänisches etymologisches 
Wörterbuch. Vol. I–II. Heidelberg: Carl Winter Universitätsbuchhandlung.
Vilim Frančić, 1961: Budowa słowotwórcza serbochorwackich kolektywów. Kraków: 
Uniwersytet Jagielloński.
Matthias Fritz, Michael Meier-Brügger, 2021: Indogermanische Sprachwissenschaft. 
Berlin, Boston: Walter de Gruyter.
Grażyna HaBrajska, 1995: Collectiva w języku polskim. Łódź: Wydawnictwo Uniwersytetu 
Łódzkiego.
Grażyna HaBrajska, Krzysztof Tomasz witczak, 1994: Słowiańskie kolektywa drzewne 
na -ьje i ich indoeuropejska geneza. Rocznik Slawistyczny 49/1. 31–39.
Grażyna HaBrajska, Mikołaj rycHło, Krzysztof Tomasz witczak, 2020: Collective 
Nouns Denoting Trees in the Scandinavian Languages. Scandinavian Philology 
18/2. 261–83.
Alexander jóHannesson, 1956: Isländisches etymologisches Wörterbuch. Bern: Francke 
Verlag.
Blaž jurišić, 1992: Nacrt hrvatske slovnice II. Tvorba imenica u povijesnom razvoju. 
Zagreb: Matica hrvatska.
Elwira kaczyńska, 2019: Laconian βοῦα ‘Band of Boys’ as a Collective Noun. Graeco-
Latina Brunensia 14/1. 93–103. 
Alwin kloekHorst, 2008: Etymological Dictionary of the Hittite Inherited Lexicon. 
Leiden, Boston: Brill. 
Hans kraHe, Wolfgang MeiD, 1967: Germanische Sprachwissenschaft. Vol. III: 
Wortbildungslehre. Berlin: Walter de Gruyter.
Bogusław kreja, 1988: Zagadnienie polskich kolektywnych nazw miejscowych typu 
Brzezie, Pustkowie. Gdańskie Studia Językoznawcze 4. 41–82.
Guus kroonen, 2013: Etymological Dictionary of Proto-Germanic. Leiden, Boston: Brill.
Roman, laskowski, 1996: Derywacja rzeczowników w dialektach laskich. Vol. I: 
Abstracta, collectiva, deminutiva, augmentativa. Wrocław, Warszawa, Kraków: 
Zakład Narodowy im. Ossolińskich.
Danuta lecH-kirstein, 2015: Nazwy drzew w śląskiej toponimii. Onomastica 59. 223–39. 
Henry George liDDell, Robert scott, 1996: A Greek-English Lexicon. Oxford: The 
Clarendon Press. 
Ranko Matasović, 2005: Collectives in -ьje in Slavic. Suvremena Lingvistika 59–60/1–2. 
35–38. 
Ranko Matasović, 2014: Slavic Nominal Word-Formation. Proto-Indo-European 
Origins and Historical Development. Heidelberg: Universitätsverlag Winter. 
Antoine Meillet, André vaillant. 1934: Le slave commun. Paris: Librairie Ancienne 
Honoré Champion.
Slavistična revija, letnik 70/2022, št. 2, april–junij172
Ewa Mianowicz, 1975. Formant -ьje w serbochorwackich toponimicznych kolektywach od 
nazw drzew. Zeszyty Naukowe Wydziału Humanistycznego Uniwersytetu Gdańskiego. 
Filologia Rosyjska 5. 55–66. 
Franco Montanari, 2018: The Brill Dictionary of Ancient Greek. Leiden, Boston: Brill.
Julius pokorny, 1959: Indogermanisches etymologisches Wörterbuch. Bern, München: 
Francke Verlag.
Stanisław rosponD, 1971: Gramatyka historyczna języka polskiego. Warszawa: 
Państwowe Wydawnictwo Naukowe.
Izabela różycka, 1989: Nazwy buk (Fagus silvatica), cis (Taxus baccata), jawor (Acer 
pseudoplatanus), modrzew (Larix polonica) i pochodne w toponimii polskiej. Acta 
Universitatis Lodziensis. Folia Linguistica 20. 53–99. 
Barbara rykiel-keMpF, 1985: Budowa słowotwórcza nazw zbiorowych w języku polskim 
XVII wieku. Wrocław, Warszawa, Kraków, Gdańsk, Łódź: Zakład Narodowy im. 
Ossolińskich. 
Franciszek sławski (ed.), 1974, 1979, 2001: Słownik prasłowiański. Tom I. Wrocław, 
Warszawa, Kraków, Gdańsk 1974; Tom III. Wrocław, Warszawa, Kraków, Gdańsk 
1979; Tom VIII. Wrocław, Warszawa, Kraków 2001: Zakład Narodowy imienia 
Ossolińskich, Wydawnictwo PAN. 
Franciszek sławski, 2011: Słowotwórstwo, słownictwo i etymologia słowiańska. Kraków: 
Polska Akademia Umiejętności. 
Wojciech sMoczyński, 2018: Lithuanian Etymological Dictionary. Vol. I–V. Berlin: 
Peter Lang Verlag. 
Ralph Lilley turner, 1966: A Comparative Dictionary of the Indo-Aryan Languages. 
London: Oxford University Press. 
André vaillant, 1964: Manuel du vieux slave. Vol. I: Grammaire. Paris: Institut 
d’Études Slaves.
Jan de Vries, 1977: Altnordisches etymologisches Wörterbuch. Leiden: Brill.
Krzysztof Tomasz witczak, 2016: Nierozpoznane collectivum w językach tocharskich: 
toch. A pratri ‘bracia’. Roczniki Humanistyczne 64/9. 119–33.
Веслав Борысь, 1979: Аб некаторых беларускiх лексічных архаізмах. Беларуская 
лiнгвiстыка 15. 47–53.
[Wiesław Boryś, 1979: Ab nekatoryh belaruskih leksičnyh arhaizmah. Belaruskaja 
linhvistika 15. 47–53.]
Анатолий Павлович НепокупНый (ed.), 1998: Общая лексика германских и балто-
-славянских языков. Киев: Наукова думка.
[Anatolij Pavlovič nepokupnyj (ed.), 1998: Obščaja leksika germanskih i balto-
-slavjanskih jazykov. Kiev: Naukova dumka.]
Олег Николаевич ТруБачев (ed.), 1974, 1977, 1990: Этимологический словарь 
славянских языков. Праславянский лексический фонд. Том I. Москва 1974; 
Том IV. Москва 1977; Том XVI. Москва 1990: Наука.
[Oleg Nikolaevič truBačev (ed.), 1974, 1977, 1990: Ètimologičeskij slovar’ slavjanskih 
jazykov. Praslavjanskij leksičeskij fond. Tom I. Moskva 1974; Tom IV. Moskva 
1977; Tom XVI. Moskva 1990: Nauka.]
173Krzysztof Tomasz Witczak, Grażyna Habrajska, Mikołaj Rychło: Proto-Slavic Collectives in *-ьje
Олег Николаевич ТруБачев, Анатолий Федорович Журавлев (eds.), 2005: 
Этимологический словарь славянских языков. Праславянский лексический 
фонд. Том XXXII. Москва: Наука. 
[Oleg Nikolaevič truBačev, Anatolij Fedorovič Žuravlev (eds.), 2005: Ètimologičeskij 
slovar’ slavjanskih jazykov. Praslavjanskij leksičeskij fond. Tom XXXII. Moskva: 
Nauka.]
Макс Фасмер, 1986: Этимологический словарь русского языка. T. II. Москва: 
Прогрес. 
[Maks vasMer, 1986: Ètimologičeskij slovar’ russkogo jazyka. T. II. Moskva: Progres.]
povzetek
V prispevku je postavljena hipoteza o indoevropskem izvoru slovanskih kolektivnih (zbir-
nih) samostalnikov na *-ьje, pri čemer je v dveh točkah v kritični pretres vzeta in nadgrajena 
interpretacija Ranka Matasovića o tej problematiki iz leta 2005. Indoevropski jeziki izkazujejo 
dva sorodna tipa kolektivov, in sicer na *-ii o̯m in *-ii e̯h2, prim. psl. *bratrьje s kol. 'bratje, 
bratovščina', ved. bhrātryám s kol. 'bratovščina', toh. A pratri 'bratje' vs. psl. *bratrьja ž kol. 
'bratje, bratovščina', at. gr. φρᾱτρίᾱ ž kol. 'pleme, rod, rodbina; fratrija, bratovščina'. Navedeni 
skupini tvorjenk morata biti obravnavani v medsebojni zvezi. Kolektivi srednjega spola so 
nedvomno izpričani v germanščini: stnord. birki s 'brezov gozd, brezje' (< pgerm. *berkijan) 
in so v zvezi s psl. *berzьje s 'brezov gozd, brezje'. Pojavljajo se tudi na zahodnobaltskem pro-
storu (prim. stprus. pannean s 'jarek, rov' = pgerm. *fanijan s kol. 'močvirje, barje'). Poleg tega 
je tovrstno gradivo znano iz stare indijščine, grščine in toharščine. Sklepna ugotovitev je, da 
praslovanski kolektivi na *-ьje niso besedotvorna inovacija (nastali niso niti iz praslovanskih 
abstraktnih samostalnikov na *-ьje niti iz indoevropskih samostalniških i-jevskih osnov). 
Povedano drugače, praslovanski kolektivi na *-ьje so indoevropska dediščina. Sklenemo 
torej lahko, da se je v severni kot tudi indijski indoevropščini pripona *-ii o̯m (s) uporabljala v 
kolektivnem (zbirnem) pomenu.