CC BY
33
DOI: 10.30842/ielcp230690152627
И. М. Егоров
Школа актуальных гуманитарных исследований РАНХиГС, Москва, Россия.
'WÖRTER' > 'SACHEN': ОБ ОДНОМ ОДНОНАПРАВЛЕННОМ СЕМАНТИЧЕСКОМ ПЕРЕХОДЕ
Настоящая статья посвящена типологическому исследованию семантического перехода существительных со значением 'слово', 'речь' и глаголов, обозначающих речевые акты, в существительные со значением 'вещь'. Свидетельства этого перехода обнаруживаются в различных индоевропейских и семитских языках, а также в таких языках, как японский, хурритский и чатино. Данный семантический переход является однонаправленным. Значение 'вещь (материальный предмет)' может развиваться из 'слово', 'речь' или из verba dicendi только через значение 'предмет (сообщения)'. Обсуждаемое семантическое изменение может иметь ареальную природу. Особенно частотно оно на древнем Ближнем Востоке.
Ключевые слова: диахроническая семантика, типология семантических переходов, индоевропейские языки, семитские языки.
Ilya M. Egorov
The School for Advanced Studies in the Humanities RANEPA. [email protected]
'Wörter' > 'S achen': a case stu dy of a unidirectional semantic shift
The present paper provides a typological study of the semantic shift from 'word' or 'speech' and from speech act verbs to nouns with the meaning 'thing'. The evidence for such a shift was found in various Indo-European and Semitic languages, as well as in Japanese, Hurrian, and Chatino. This semantic shift is unidirectional. The meaning 'thing (material object)' can be developed from 'word', 'speech' or verba dicendi only through the stage 'matter'. The semantic change in question can have an areal nature. Ancient Near East is an area where it is especially common.
Keywords: diachronic semantics, typology of semantic shifts, Indo-European, Semitic.
1. Introduction
In Goethe's tragedy, Faust step by step comes to the understanding of Лоуод as 'deed' instead of commonly accepted 'word' rethinking the translation of the Gospel of John. Observations under the natural languages show that similar semantic developments are not uncommon. Lexemes for 'word / speech' or some kind of speech act verbs (I will further collectively refer to
them as to a speech-domain) often yield such meanings as 'affair', 'deed', 'matter', and finally 'thing'. The aim of the present paper is to give a typological overview of this somewhat paradoxical semantic shift and to find a reason why it goes exactly in this direction and almost never in the reverse one.
The traditional approach to the typology of semantic shifts was concentrated on the main kinds of changes such as widening, narrowing, metaphor, metonymy, synecdoche etc. (see Urban 2015: 374-377). A new trend in this field is represented by empirical-based studies of tendencies or even universals in the semantic development of lexemes within certain semantic domains. Among the most important early studies of this type must be mentioned Nicholas Evans and David Wilkins' seminal article (2000) which introduces an explanatory model involving the notion of bridging contexts.
The appearance of such databases as CLICS (Rzymski et al. 2019) and DatSemShift (Anna Zalizniak et al. 2012) became another milestone. The data for the present study comes mostly from these databases. In the second step of data collection, the material was verified and extended with synchronic dictionaries and corpus examples if available. Unfortunately, the quality of lexicographic descriptions is not always sufficiently high, so in some cases, our knowledge of semantics is limited to brief glosses. Mapping of the semantic field in question must be conducted before the investigation into its diachronic typology. I follow here Alexandre François' approach (2008) based on colexification patterns. After the clarification of synchronic semantics, etymological analysis was undertaken. I avoid proposing new etymologies and rely on the existing expert views. The direction of the semantic shifts is not always explicated in etymological dictionaries. In most such cases, it can be set based on the distribution of meaning between languages. Explaining why this semantic change is possible and why it goes exactly in this direction I rely on Evans & Wilkins 2000.
The notion of semantic shift (= semantic change) is understood in the present paper in the traditional way, i.e. as a diachronic event. Such understanding was dismissed in the influential Anna Zaliz-nyak's database of semantic shifts, where the term shift is used for synchronic polysemy, i.e. Zaliznyak treats synchronic polysemy as a particular case of a diachronic change. It seems reasonable to rethink her "types of realization of semantic shifts" (see Zalizniak et al.
2012: 634-635) in the following way. Synchronic polysemy is actually one of the ways the result of changes in meaning can be observed. In rare cases, it can be observed in written sources, sometimes the shift is assumed based on the meaning of cognate words in related languages. The change in semantics can occur by itself or can be accompanied by derivation. If the connection between a primary stem and a derivative disappears one can speak about purely etymological relatedness.
2. Mapping of the semantic field thing
The English word thing is problematic since it covers several semantic slots which can be filled by different lexemes in other languages. The most plausibly distinguishable separate slot is 'personal things, property, possessions'. For example, Kazym Khanty purmas is used exactly in this meaning whereas ut refers to some unspecified material or abstract objects. Purmas is a common everyday word, not a special legal term. A similar situation is observed in Turkish, where the slot 'property' is occupied by e§ya, whereas 'an unspecified material object' is denoted with nesne. Nesne tends to be limited to material objects. For the reference to abstract notions and situations, the lexeme §ey is used (applicable for material objects as well). The meaning 'situation, deed, (state of) affair' is clearly different from the meaning 'an unspecified abstract notion' (such as love, truth, faith etc.). Probably most languages have some specific words for the former meaning. However, these semantic slots can be distinguished in the scope of more abstract lexemes from thing-domain as well, cf. the following Russian examples, where the use of the word predmet is limited to 'abstract objects':
(1) a. ljubov' eto sloznaja vesc' / sloznyj predmet... love it complicated thing / complicated thing 'Love is a complicated thing...'
b. tam proizosli strasnyje vesci / *predmety
there happen terrible things / things 'The terrible things have happened there'
Turkish provides a reason to distinguish one more slot in the thingdomain which can be labeled as 'matter', i.e. discussed things
(Turkish konu). Summing up, one can subdivide the semantic field thing into five slots as shown in Table 1.
Table 1. Structure of the semantic field thing
Semantic slot Typical context Kazym Khanty Turkish Russian
'property' I haven't touched your things. purm3s e?ya vesc'
'material object' I don't understand what this thing is used for. ut nesne §ey vesc' predmet
'abstract notion' Love is a complicated thing. ut vesc' predmet
'(state of) affair' I didn 't know that such a thing had happened. ut e vesc'
'matter' I don't want to discuss this thing with you. ut konu vesc' predmet
These slots can be combined into two groups. One of them refers to some substances ('property' and 'material object') and another one to non-substances ('abstract notion', '(state of) affair', 'matter'). The distinction of the particular meanings inside these groups often is not so clear-cut.
3. Evidence for the shift from speech-domain to thing-domain
3.1. Slavic *vektb and Germanic *wehti-
Slavic and Germanic demonstrate nominal formations from PIE *wekw- 'say' (LIV2: 673). PIE *wekw-t-i- > PGerm. *wehti- > Goth. waihts 'thing, entity, matter', ON vettr 'being, thing', OE wiht 'being, demon, thing', OSax. wiht 'something', OHG wiht 'being, demon, thing' (Kroonen 2013: 578); PSlav. *vektb > Czech vec, Sloven vec, USorb. wjec, Bulg. вещ, OCS vestb (Machek 1968: 680; Georgiev et al. 1971: 140). Russian vesc' must be treated as a Bulgarian loan due to the sc instead of c < *kt expected in East Slavic (Vasmer 1953: 196). In most of the ancient Germanic languages, the reflex of *wehti- became a place-holder (frequently used with negation) filling all the slots in the thing-domain. Its semantics differs in minor details among the Germanic languages. The Slavic lexemes cover the whole thing-domain as well.
Guus Kroonen (2013: 578) suggests PIE *weg-t-i- and devoicing of the palatal guttural in Slavic. Such a solution meets both formal and semantic problems. Firstly, the expected reflex of PIE *k is PSlav. *s. Although some "centum reflexes" are known in Slavic, I see no reason to postulate one more formation with such an irregular development. If the root *weg- suggested by Kroonen had been the same item that the stem *weg- 'munter, lebhaft, kräftig' (LIV2: 660661), it would be difficult to explain the semantic development. The reconstruction *wekw-t-i- is unproblematic from the phonological point of view, since PGerm. *ht is a regular reflex of the PIE cluster *kwt, cf. PIE *nokw-t- > PGerm. *naht-. The semantic shift from a speech act verb to 'thing' is quite common as will be seen from further consideration.
3.2. Classical Armenian ban
PIE *bhehrn-i- > Classical Armenian ban 'speech, word; Logos; thing; precept, commandment, etc.'. Overview of the contexts in the Biblical corpus shows that it covers predominantly the slot 'matter' in the thing-domain. It is also attested with the meaning 'work, business; thing; subject' in the modern dialects. The PIE *№eh2- is a well-known root reconstructed based on Gr. yn^i 'speak, say', ORus. bajati 'talk', Ved. bhanati 'speak, sound' (LIV2: 69). Other notable reflexes of this stem are the Classical Armenian verb bam 'speak, say' and dialectal banel 'to work' (Martirosyan 2009: 166).
3.3. Tocharian B wäntare
PIE *wnd-rwo- > Toch. B wäntare 'thing, affair, happening, object, matter' (Adams 2013: 643). The lexeme covers all slots in the thing-domain, except for 'property'. This formation has the following Indo-Iranian comparanda: Skt. vandaru- 'praising, praise', Av. duz-uuandru- 'blaspheming' (Mayrhofer 1996: 502-503). It should go back further to the PIE *wend- 'speak (solemnly)' (LIV2: 681). The connection with Toch. A, B we-n- 'speak, say, state, tell' is doubtful as well as the etymology of this verb. This verb can originate either from *wend- (LIV2: 681), or *wekw- (LIV2: 673), or *hjwed(H)- (LIV2: 286), see Adams 2013: 659-660 and Malzahn 2010: 910 for discussion.
3.4. Hittite uttar, Hittite memiya(n)-
These two words are distributed chronologically; uttar is more archaic. As noted in Guterbock et al. 1980: 274, memiya(n)- appears instead of uttar in some younger copies of older texts. Both lexemes refer to non-substances. An indefinite pronoun is typically used for reference to unspecified material objects; the meaning 'property' is expressed with assu-.
PIE *weth2-r- / *uth2-en- > Hitt. uttar / uddan- 'word, speech; thing, case; story; reason'. Here I accept Kloekhorst's etymology (2008: 933), who assumes this word contains the PIE root *weth2-'say' (LIV2: 694). It must be noted that alternative solutions suggest the connection with verbs denoting speech acts as well. Eichner (1980: 146 footnote 69) connects this stem with *h2wedH- (LIV2: 286); Rieken (1999: 301) proposes a relation to an ad hoc reconstructed root *h1ew- 'speak'. The cognate word utar 'word; spell' is attested in Cuneiform Luvian (Melchert 1993: 247).
Hitt. memiya(n)- is a synchronic derivative from mema-, the basic verb ' say' in New Hittite, cf. Puhvel's gloss (Puhvel 2004: 126-140): 'speak, speak of, say, tell, utter, pronounce, mention, declare, report, read aloud, recite; bespeak, promise', see also Guterbock et al. 1980: 254-263.
3.5. Old High German redina
OHG redina is a synchronic derivative from the verb redinon 'say, tell', cf. also the verb redion 'tell, say, report' and the noun reda 'message, story'. These words go back to PGerm. *rapjon-, cf. Goth. rapjo 'counting, account, explanation, number', OFris. rethia 'talk, account', OSax. rethia 'talk, account'. The survey of available corpus examples shows that the word is used for 'matter' and hardly occurs with the meaning '(state of) affair' outside the context of narration.
3.6. Polish rzecz, Slovene rec
The etymology of Polish rzecz 'thing' and Slovene rec 'thing' clearly points to the innovative nature of this meaning. The Polish and Slovene words cover the whole thing-domain; in Slovene, it competes with stvar. The nouns represent the lengthened grade of the PSlav. verb *rekti 'say, speak' (see LIV2: 506), cf. also the meanings of the Slavic cognates: Czech rec 'speech', Slovak rec 'speech', USorb. and LSorb. rec 'speech, language', Ukr. ric
'thing', Bel. rec 'thing', ORus., OCS recb as an equivalent for Gr. ptf^a, Xoyoq, Siavoia, Bulg. rec 'word', SCB rijec 'word', Slovene rec 'thing'. Belarusian and Ukrainian forms show the results of the Polish influence. The semantic development in Polish and Slovene must be independent of each other.
3.7. Akkadian amatu
Akkadian amatu is glossed in the following way in (CAD 1968: 29): 'spoken word, utterance, formula; news, report, message, rumor, secret, interpretation, plan thought; wording, text, content, terms of an agreement' 'command, order, decision; legal case, case in court, legal transaction; matter, affair, thing'. Albrecht Goetze (1947: 245) connects this noun to the verb amu and compares it with Ugaritic hwt 'word'.
3.8. Old Hebrew dabar
Old Hebrew dabar is a synchronic derivative from the root dbr 'speak'. The noun demonstrates synchronic polysemy 'word / matter / thing' (HALOT: 211). Studied contexts show that it covers predominantly the slots 'matter' and '(state of) affair'. The polysemy can be nicely illustrated with the example from Gn. 15:1:
(2) ?ahar ha-ddab arim ha-?elleh hayah dab ar Yahweh after det-thing det-this come word Yahweh ?el-?abram ba-mmahzeh to-Abram in-vision 'After these things the word of the Lord came unto Abram in a vision' (King James' Bible translation)
3.9. Ge'ez nagar
Ge'ez noun nagar is a synchronic derivative from the verb nagara 'say, tell, speak, talk, relate, recite, proclaim, declare, report, announce, indicate, inform'. The noun is glossed by Wolf Leslau (1987: 392) in the following way: 'speech, talk, word, language, saying, pronouncement, discourse, statement, thing, affair, subject, account, matter, situation'. He connects the root to Akk. nagaru 'notify' and Ug. ngr 'herald'. Modern Ethiosemitic languages preserve the polysemy 'word / matter'.
3.10. Ge'ez qal
Another Ge'ez noun demonstrating synchronic polysemy is qal 'voice, word, saying, speech, statement, discourse, command, order, sound, noise, expression, maxim, thing' (Leslau 1987: 426). Survey of the Biblical corpus shows that this word occupies the slots 'matter' and '(state of) affair' but clearly not 'material object' and 'property'. Such cognates as Arab. qala 'say', Heb. qol 'voice' clearly indicate the original meaning belonging to the speechdomain. Among the modern Ethiosemitic languages, Gurage preserves polysemy 'voice, thing'; cognates in Tigre, Tigrinya, and Amharic mean 'word'.
3.11. Soqotri btt'e
Soqotri bil'e 'thing' (Naumkin & Kogan 2018: 431) represents the result of the semantic shift 'thing' < 'word'. The original semantics can be reconstructed based on the cognates from closely related Modern South Arabian languages: Mehri bdhlit/bdhel 'word', Jibbali behlet/behdl 'word' (Johnstone 2006: 45; Johnstone 1981: 24). Antiquity of the Mehri and Jibbali meaning is proven with the data from the other more distantly related Semitic languages: Ge'ez bahla 'say, speak, call, announce, command'; Arab. bahala 'pronounce a curse', Akk. ba'alu, balu 'beseech, pray'(Leslau 1987: 89).
3.12. Hurrian tiwe
Hurrian tiwe demonstrates synchronic polysemy 'word / thing', once attested with the meaning 'name'. It corresponds to Akk. amatu and Hitt. uttar, memiya(n)- according to Laroche 1980: 267268. The word is derived from the verb tiw- 'say'. Urartian cognates: ti=ni 'name', ti=a 'say' (Richter 2012: 453; Kassian 2011: 406, 413).
3.13. Japanese1
Japanese koto fills the slots 'matter' and '(state of) affair'. It can have the meaning 'word; remark; statement' as well. Polysemy apparently must be reconstructed onto the Proto-Japonic level. However, the range of derivatives (kotobo 'word', kotobuki
1 I am grateful to George Starostin for pointing me to this example.
'congratulations', kotodate 'announcement, pronouncement', kotodomori 'stammering') could speak for the antiquity of the meaning in the speech-domain. This lexeme is opposed to mono denoting 'material object' or 'property'.
3.14. Other examples
Four further examples of the synchronic polysemy 'word / thing' were found. However, semantic and etymological details remain unknown due to the scarcity of existing sources:
Chatino (Otomanquean) cha' 'word; sentence, speech; reason; thing' (Pride & Pride 2010: 7-8);
Konyak (Brahmaputran < Sino-Tibetan) ngao 'word; thing'; Lushai (Kuki-Chin-Naga < Sino-Tibetan): thu 'noun: word, saying, news, description, account, statement, narrative, history, information, order, command, authority, will, wish, pleasure, affair, business, concern, subject, thing, things, cause, case; preposition: about, regarding, concerning, of (Lorrain 1975: 485); Sangtam (Kuki-Chin-Naga < Sino-Tibetan):yu 'word; thing'.
4. Explaining the semantic shift
A lot of words that preserve a synchronic connection to the speech-domain do not occupy the whole thing-domain, but only the slots implying reference to non-substances. The lexemes referring to both substances and non-substances preserve only etymological relation to words from the speech-domain. One can speculate that the latter group of lexemes represents a more advanced stage, and assume the following scenario. A word from the speech-domain first develops the meaning 'matter' and then it extends to 'material object' while losing its original meaning or transparent derivational structure.
The proximity of the meaning 'matter' to speech-domain can easily be proven with bridging contexts. I understand here the bridging context as a case of a contextual polysemy following Evans & Wilkins 2000: 550. The starting point for the studied semantic shifts is such contexts where the lexeme can denote simultaneously a discourse in its linguistic form (words, speeches) and a discussed matter (situation), so it is a metonymic shift. Such sentences are relatively frequent in the written sources, cf., e.g., the Old High German and Hittite examples:
(3) Otfrid II.9.1
thisu selba redina the ih zalta this same thing that I tell 'All the things/words I told (about)'
(4) HKM 32 M§t. 75/117 cited after Hoffner 2009: 158 uddar=mu kue hatraes nu=at ASME word=1sg.dat that write and=acc.pl.n I.hear
'I have heard the words that you wrote to me' ~ 'I have heard about the matter that you wrote to me'
The development 'matter' > 'thing' is attested also without special connection to the speech-domain. The Germanic languages give the most obvious examples: PGerm. *pinga- > OE ping 'meeting, case, thing' > E thing; OHG ding 'meeting, court, thing, affair' > G Ding 'thing' (Kroonen 2013: 542). Somewhat similar development from a kind of legal term can be illustrated with another German example and Romance ones: PGerm. *sako- > OHG sahha 'case, guilt', G Sache 'affair, matter, thing'; Latin causa 'judicial process, lawsuit, case' > Fr. chose 'thing', Sp. cosa 'thing'.
5. Areal distribution
Patterns of colexification frequently spread within some linguistic areas and appear in genetically unrelated (or not closely related) languages. See the following recent works on the areal nature of colexification: Koptjevskaja-Tamm & Liljegren 2017; Zhivlov 2019; Segerer & Vanhove 2021; Georgakopoulos et al. 2021. Concerning the data discussed above, one can speak about Ancient Near East as an area characterized by the semantic shift in question. It is manifested in Akkadian, Hittite, Hurrian, and Old Hebrew. All these languages demonstrate some other outcomes of language contact, namely Akkadian borrowings, Hurrian borrowings in Hittite, a number of shared Wanderworter. Cultural contacts between the speakers of these ancient languages are well documented. These are perfect conditions for the spread of a colexification pattern.
Biblical translations probably played some role in the spreading of the polysemy 'word / thing' or at least 'word / matter'. Notable is the use of Ge'ez nagar and Classical Armenian ban (primary
meanings of both lexemes clearly belong to speech-domain) in place of Old Hebrew dabar and Greek p^p.a:
(5) a. Old Hebrew
ha-yippale Yahweh dabar
qst-be.too.hard Yahweh thing
b. Greek
fiy dSvvazei napa z& Gem pqpa is impossible to the God thing
c. Ge'ez
bonu nagar za-yasa?nu l-?egzi?abher
is.there thing that-be.impossible for-God
d. Classical Armenian
mit e tkaranayce ar i yAstucoy ban
is.it.possible weaken by to God thing
'Is any thing too hard for the Lord?' (Gn. 18:14, King James' Bible
translation)
Although the translations of the Bible were crucial for the earlier stages of Armenian and Ethiopian, one can only speculate whether it was the source for the development of semantic shift 'word' > 'matter / thing'.
6. Conclusions
The results of the semantic shift from speech- to thing-domain are found in the number of Indo-European, Semitic, Sino-Tibetan languages as well as in Japanese, Chatino, and Hurrian. Ancient Near East can be posited as an area the shift in question was typical for. The transitional phase of the shift is represented with bridging contexts allowing both interpretations 'speech / words' and 'matter (of the speech)'. The meaning 'material objects' can be developed only through the stage of the meaning 'matter'.
Abbreviations
Languages
Akk. — Akkadian; Arab. — Arabic; Av. — Avestan; Bel. — Belarusian; Bulg. — Bulgarian; E. — English; Fr. — French, G. — German; Goth. -Gothic; Gr. — Greek, Hitt. — Hittite; LSorb. — Lower Sorbian, OCS. — Old Church Slavonic; OE. — Old English; OFris. — Old Frisian; OHG — Old High German; ON — Old Norse; ORus. — Old Russian; OSax. — Old Saxon; PGerm. — Proto-Germanic; PIE — Proto-Indo-European; PSlav.
— Proto-Slavic; SCB — Serbian, Croatian, Bosnian; Skt. — Sanskrit; Sp.
— Spanish; Toch. A — Tocharian A; Toch. B — Tocharian B; Ug. — Ugaritic; Ukr. — Ukrainian; USorb. — Upper Sorbian; Ved. — Vedic Sanscrit.
Sources
CAD — Civil et al. 1968.
HALOT — Koehler & Baumgartner 1994.
LIV2 — Rix et al. 2001.
References
Adams, D. Q. 2013: A dictionary of Tocharian B (Leiden Studies in Indo-
European 10). 2nd edn. Amsterdam, New York: Rodopi. Civil, M., Gelb, I. J., Landsberger, B., Oppenheim, A. L., Reiner E. 1968: The Assyrrian dictionary of the oriental institute of the University of Chicago. Vol. 1 Part 2. A. Chicago: The Oriental Institute. Eichner, H. 1980: Phonetik und Lautgesetze des Hethitischen - ein Weg zu ihrer Entschlüsselun. In: Manfred Mayrhofer (ed.). Akten der VI. Fachtagung der Indogermanischen Gesellschaft, 120-165. Wiesbaden.
Evans, N., Wilkins, D. 2000: In the Mind's Ear: The Semantic Extensions of Perception Verbs in Australian Languages. Language 76 (3). 546-592.
François, A. 2008: Semantic maps and the typology of colexification: Intertwining polysemous networks across languages. In: Martine Vanhove (ed.). Studies in Language Companion Series. Vol. 106, 163-215. Amsterdam: John Benjamins Publishing Company. doi:10.1075/slcs.106.09fra. Georgakopoulos, T., Grossman, E., Nikolaev, D., Polis, S. 2021: Universal and macro-areal patterns in the lexicon: A case-study in the perception-cognition domain. Linguistic Typology. 000010151520212088. doi:10.1515/lingty-2021-2088. Georgiev, V. I., Gäläbov, I., Zaimov, Y., Ilcev, S. 1971: Bälgarski etimologicen recnik [Bulgarian etymological dictionary]. Vol. 1. Sofia: Izdatelstvo na Bälgarskata Akademija na Naukite. Goetze, A. 1947: Short or Long a? (Notes on Some Akkadian Words).
Orientalia Nova Series 2(16). 239-250. Güterbock, H. G., Hoffner, H. A., van den Hout, T. P. J., Goedegebuure, P. M. (eds.). 1980: The Hittite Dictionary of the Oriental Institute of the University of Chicago. Vol. L-N. Chicago: University of Chicago Oriental Institute. Hoffner, H. A. 2009: Letters from Hittite Kingdom (Writings from the
Ancient World 15). Atlanta: Society of Biblical Literature. Johnstone, T. M. 1981: Jibbäli lexicon. New York: Oxford University Press.
Johnstone, T. M. 2006: Mehri lexicon. London; New York: Routledge. Kassian, A. 2011: Hurro-Urartian from the lexicostatistical viewpoint.
Ugarit-Forschungen (42), 383-452.
Kloekhorst, A. 2008: Etymological dictionary of the Hittite inherited lexicon (Leiden Indo-European Etymological Dictionary Series). Leiden: Brill.
Koehler, L., Baumgartner, W. 1994: The Hebrew and Aramaic lexicon of the Old Testament. Leiden; New York; Köln: BRILL.
Koptjevskaja-Tamm, M., Liljegren, H. 2017: Semantic Patterns from an Areal Perspective. In Raymond Hickey (ed.), The Cambridge Handbook of Areal Linguistics, 204-236. Cambridge: Cambridge University Press. doi:10.1017/9781107279872.009. https://www.cambridge.org/core/product/identifier/9781107279872 %23CN-bp-8/type/book_part (17 March, 2022).
Kroonen, G. 2013: Etymological dictionary of Proto-Germanic (Leiden Indo-European Etymological Dictionary Series 11). Leiden: Brill.
Laroche, E. 1980: Glossaire de la langue hourrite. Paris: Éditions Klincksieck.
Leslau, W. 1987: Comparative dictionary of Geez (Classical Ethiopic). Wiesbaden: Otto Harrassowitz.
Lorrain, J.. 1975: Dictionary of the Lushai language. Calcutta: Tha Asiatic Society.
Machek, V. 1968: Etymologicky slovník jazyka ceského [Etymological dictionary of Czech language]. 2nd edn. Praha: Ceskoslovenská akademie ved.
Malzahn, M. 2010: The Tocharian verbal system (Brill's Studies in Indo-European Languages & Linguistics). Vol. 3. Leiden, Boston: BRILL.
Martirosyan, H. M. 2009: Etymological dictionary of the Armenian inherited lexicon (Leiden Indo-European Etymological Dictionary Series 8). Leiden, Boston: BRILL.
Melchert, H. C. 1993: Cuneiform Luvian lexicon. Chapel Hill, N.C.
Naumkin, V., Kogan, L. (eds.). 2018: Corpus of Soqotri Oral Literature (Studies in Semitic Languages and Linguistics 95). Vol. 2. Leiden, Boston: BRILL.
Pride, L., Pride, K. 2010: Diccionario Chatino de la Zona Alta: Panixtlahuaca, Oaxaca y otros pueblos (Serie de Vocabularios y Diccionarios Indígenas "Mariano Silva y Aceves" 47). México: Instituto Lingüístico de Verano.
Puhvel, J. 2004: Hittite etymological dictionary (Trends in Linguistics. Documentation 22). Vol. 6: Words beginning with M. Berlin; New York: Mouton de Gruyter.
Richter, T. 2012: Bibliographisches Glossar des Hurritischen. Wiesbaden: Harrassowitz Verlag.
Rieken, E. 1999: Untersuchungen zur nominalen Stammbildung des Hethitischen (Studien Zu Den Bogazköy-Texten 44). Wiesbaden.
Rix, H., Kümmel, M. J., Zehnder, T., Lipp, R., Schirmer, B. 2001: Lexikon
der indogermanischen Verben: die Wurzeln und ihre Primärstammbildungen. 2nd edn. Wiesbaden: Dr. Ludwig Reichert.
Rzymski, C., Tresoldi, T., List, J.-M., Greenhill, S., Forkel, R. 2019: The Database of Cross-Linguistic Colexifications, reproducible analysis of cross-linguistic polysemies. https://clics.clld.org/.
Segerer, G., Vanhove, M. 2021: Areal patterns and colexifications of colour terms in the languages of Africa. Linguistic Typology 0 (0). 000010151520212085. doi:10.1515/lingty-2021-2085.
Vasmer, M. 1953: Russisches etymologisches Wörterbuch. Vol. 1 A-K. 3 vols. Heidelberg: Carl Winter.
Zalizniak, Anna A., Bulakh, M. S., Ganenkov, D. S., Gruntov, I. A., Maisak, T. A., Russo, M. M. 2012: The catalogue of semantic shifts as a database for lexical semantic typology. Linguistics 50(3), 633-669.
Zhivlov, M. 2019: Areal polysemy 'earth/year' in North American languages: historical implications. Etnografia 5(3). 167-180. doi:10.31250/2618-8600-2019-3(5)-167-180.
