(A shortened version of Szeverényi Sándor 2014: On the Project of a Diachronic Cognitive Onomasiological Dictionary of the Nganasan Language. Finnisch-Ugrische Mitteilungen 38: 271-287.)

1. Introduction

The aim of the article is to give an overview of the theoretical background of the project of  the diachronic cognitive onomasiological online dictionary of the Samoyedic languages. The main focus is to compile the Nganasan section in the period of 2012˗2015. In order to create this dictionary, a specific structure and program needs to be set up. Since the relations of the word forms and their meanings are classified by formal and semantic/cognitive points of view, the existing diachronic and synchronic lexicographical programmes do not relate to this project. These classifications allow different and complex search and browse capabilities, which will provide an easy access to the dictionary. Developing the section of Nganasan is perfectly realistic, and would create a ”historical thesaurus” of the Nganasan language. Once the Lexicographical Program is created, the database can be extended in several directions (e.g. as regards the number of the languages, lexemes, concepts, semantic domains, analytical parts, etc.). The classification of the relations between lexemes will require basic research such as etymological, and areal linguistic study. The systemizing of the classifications could result in new discoveries relating to the modeling of lexical pathways. 

2. The Nganasan language 

The main focus of the research lies on the Nganasan language with the following supporting arguments: 

(1) The Nganasan electronic corpus exists. When it comes to building corpora, projects utilizing less documented languages are not only costly, but time-consuming, as well. The advantage of my project is that I am able to utilize the electronic corpora of Nganasan provided in the last 10-15 years. I was fortunate to participate in some previous projects, which consequently gave me the opportunity to jumpstart my research. The most important items of the electronic corpora are: 

A) Dictionaries 

The Morphological Dictionary of the Nganasan Language (Wagner-Nagy 2006). Cc. 6000 entries. 

Nganasan-Russian-Nganasan dictionary by Kosterkina–Momde–Zhdanova (2001). Cc. 8000 entries. 

B) Texts 

Kazys Labanauskas’s folklore text collection (2001) – analyzed in our previous project (The Computational Morphological Analyzer of the Nganasan Language, OTKA). Cc. 3 500 sentences. Materials of Valentin Gusev’s research group from 2005-2008. Cc. 50,000 sentences.

C) Own materials (collected in 2008 with Beáta Wagner-Nagy in Ust-Awam), mostly lexicological questionnaires and texts (Cc. 1000 sentences). 

3. Why the need for a diachronic cognitive onomasiological  dictionary?

Similar historical and electronic databases do not contain searchable classifications of the formal and cognitive relations such as the Historical Thesaurus of English (University of Glasgow, libra.englang.arts.gla.ac.uk/historicalthesaurus). In my dissertation (Szeverényi 2008) I applied methods of diachronic cognitive onomasiology with emphasis on the property concepts. This dictionary can be applied to numerous linguistic subdisciplines such as diachronic onomasiology, and semasiology. In the last 20 years the traditional onomasiology got fresh impulse by the development of cognitive semantics. This statement is shown in the foundation of Onomasiology Online (by Joachim Grzega). There is very important research from the DECOLAR project [University of Tubingen, SFB441: Linguistic Data Structures: On the Relation between Data and Theory in Linguistics, 1999–2008 – first of all by Peter Koch and Andreas Blank, B6 project: Lexikalische Motivation im Französischen, Italienischen und Deutschen (LexiTypeSyn)], Dirk Geeraerts’s works (Geeraerts 1997, 2010), research on the English historical lexicology (Christian Kay, especially Kathryn Allan 2007),. Koch and Blank emphasize in many places that the diachronic cognitive onomasiology has numerous advantages in contrast to the traditional etymological researches. Namely, ”it enables us to discover material that is interesting independently of any etymological relationship. In this way we can postulate a potentially polygenetic evolution within one and the same language family” (Koch 2008: 109). 

The applied method is the diachronic model of lexical ‘filiation‘ (Gévaudan – Wiebel 2004, Gévaudan 2007), which can be considered as the diachronic model for lexical typology. Essentially it implies investigation of lexicalizational processes, classifying and modelling of the processes, however, as Joachim Grzega mentions DCO „is more or less the theoretical side of the practical combination of semasiology, onomasiology, word formation, and etymology” (Grzega 2002: 1023). The newness of this approach actually lies in the combination of these areas as a multiple approach through an easily readable access and reasonable system on computer. It takes into consideration all kinds of lexical evolutions and offers standardised explanations of all different kinds of lexical innovation (Gévauadan – Wiebel 2004: 2). 

The basic thesis of the model is that „formal and cognitive motivation are two dimensions of non-phonetic linguistic motivation, since there is no formal motivation without cognitive motivation and vice versa” (Koch – Marzo 2007: 262). From a typological point of view, Koch and Marzo emphasize three major scopes for its application (Koch – Marzo 2007: 273):

(i) motivational profiles of particular languages;

(ii) identification of cross-linguistic motivational tendencies and idiosyncrasies with respect to the language facts (are there more or less formally transparent languages? Are there predominantly metaphorical languages? etc.)

(iii) universal/language type-specific motivational preferences and gaps with respect to combinations in the two-dimensional grid

The investigation of a lexical motivational mechanism has been worked out a two and a three dimensional model in Tübingen. Gévaudan‘s filiation model (2007) has been applied e.g. in DECOLAR project (http://www.decolar.uni-tuebingen.de/; Bienvenus sur le site du Dictionnaire Étymologique et Cognitif des Langues Romanes). The two dimensional grid is applied to synchronic motivation investigations. Gévaudan‘s three dimensional filiation model has three components:

1. semantic filiation (Semantische Filiation): determination of cognitive relations between related lexical units;

2. morphological filiation (Morphologische Filiation): determination of formal relations between related lexical units;

3. stratificational filiation (Stratische Filiation): comparison of language strata. There are two subgroups: direct and paradigmatic filiation (Stratisch-direkte Filiation – Stratisch-paradigmatische Filiation). E.g. borrowing belongs to the former group, lexical continuity (Stratische Kontinuität), antonomasy, onomatopoezis etc. belong to the second one as calque, analogy etc.

Besides the diachronic onomasiology there are two fields playing important roles in the construction of the dictionary :

Etymological research 

Since the publishing of Janhunen’s Samoyedic Etymological Dictionary (1977), the published Samoyedic materials have been multiplied; however, these materials have been analyzed diachronicly only partially (Helimski 1986, 1992-93, 1997; Aikio 2002). This dictionary could be the first step towards the building of a historical dictionary of the Samoyedic languages. 

It is good to see that as results of descriptive fieldworks some new historical-comparative dictionaries have recently been edited, such as Irina Nikolaeva’s Yukaghir Historical Dictionary (2006), Comparative Eskimo Dictionary with Aleut Cognates (1994, 2010, by Michael Fortescue et al.), Comparative Chukotko- Kamchatkan Dictionary(2005, Michael Fortescue), Heinrich Werner’s dictionary (2002), the Relationship of Nivkh to Chukotko-Kamchatkan(2011, Michael Fortescue). Hence, it seems that today the focus is on the history of individual languages and language families as opposed to „large-scale” Siberian theories. 

Cognitive metaphor-theory 

The results of the cognitive metaphor researches (e.g. Lakoff–Johnson 1980; Kövecses 2005) can be applied in lexicology as well. It is very important because data from Siberian languages have not been used as sources of the cognitive metaphor-theory. However, it must be mentioned that research on Nganasan texts, from a figuratively point of view, are limited. As a result, the only area for improvement is in lexicology. Additionally, diachronic metaphor research has shown new results as well (e.g. HTE online dictionary, Kathryn Allan’s monograph, 2007). 

4. The structure of the dictionary

The primary goal of the dictionary is to classify the cognitive and formal relations of the lexemes and concepts, and to systematize and interpret the classification. The starting point is the well-known motivational square

lexeme 1 (L1) concept 1 (C1) 

lexeme 2 (L2) concept 2 (C2) 

(L1 = source form, L2 = target form, C1 = source concept, C2 = target concept)

Where C1 is in cognitive relation with C2, L1 is in formal relation with L2. We must classify both the cognitive relation (C1>C2) and the formal relation (L1>L2), then systematize the classifications of concepts and forms. At last we will classify the pathways, i.e. the kind of tendencies (universal, culture-specific etc.) that can be established. Problems may arise during my analysis, and below are some preceding statements: 

• Source concept/source forms can be target concept/target forms at the same time. This means that theoretically never-ending semantic chains can be detected. 

• One meaning (concept) belongs to one form. 

• The reconstructed protoforms are labeled as absolute source form and source concept. 

• The direction of the semantic change is not always clear. In such cases, fictive stems will be established as a collective source. 

To illustrate the schema, I show two examples:

forms: L1: basa >> semantic change >> L2: basa

concepts: C1: IRON, METAL >> contiguity >> C2: MONEY

   (Material for Object)

forms: L1: ŋǝnduj >> compounding >> L2: tuu ŋǝnduj

concepts: C1: A KIND OF BOAT >> similarity >> C2: STEAMBOAT, STEAMSHIP

            >> taxonomic subordination >>  

(Tuu is the genitive form of tuj ‚fire‘.)

According to the motivational square, the following informations of a lexeme should be contained: 

(1) Lexeme (form) 

(1a) language 

(1b) parts of speech 

(1c) morphological structure 

(1d) opacity [+/–] or [?] 

(1e) literature (sources) 

(1f) in context (if exists) 

(1g) comments (e.g. frequency of the forms; certainty of the etymology etc.). 

(2) concepts (< semantic domains) 

(3) pathways: 

(3a) source forms of 

(3b) source concepts of 

(3b/I) classification of the cognitive link 

-Main types of the relations: contiguity, similarity, partiality, contrast 

(3b/II) Classification of formal relation of the two forms (source and target): 

compound, derivation, conversion, loan etc. 

(3c) target form(s) of 

(3d) target concept(s) of 

(3d/I) classification of the cognitive link 

Main types of the relations: contiguity, similarity, partiality, contrast 

(3d/II) labeling of the source and target forms 

Classification of formal relation of the two forms (source and target): 

compound, derivation, conversion, loan etc. 

In the case of classification (cognitive relations and semantic domains) a good and adequate system can be found in the literature. We analyze the cognitive relations (and cognitive motivations) according to the works of Grzega (2004), Blank (2001), and Koch (2001). The basis of the classification of the forces (processes) is Grzega’s work (2004), and Blank’s work (2001). The system of the semantic domains of the Nganasan language can be created based on these systems. The basis of the system of the semantic domains is the Dictionary Development Process (http://www.sil.org/computing/ddp/DDP_downloads_tb.htm) for the description of underdocumented languages (List of Semantic Domains (v.4) for Rapid Word Collection (http://rapidwords.net/). However, I want to apply other systems too (e.g. Tucker 1998 that I used in my PhD-dissertation for the property concept), although the Zhdanova–Momde–Kosterkina’s Nganasan dictionary serves as a base for the semantic domains of Nganasan. Besides classification, the most important feature of the dictionary is the complex search and browse. Here are some examples for multiple searches with the five main components, i.e. language / lexeme / concept-domain / relation / process): 

A) Nganasan forms where the source concepts belong to COLOR 

B) Cognitive relations where the formal process is caritive derivation 

C) Metaphorical relations where the target form is X 

D) Conceptualization of a concept in different Samoyedic languages 

4.4. Problems

4.4.1 Languages and dialects

The Nganasan language has only two dialects: Avam and Vadeev, which do not differ significantly from each other. The dictionary in the case of Nganasan should not be concerned with dialects due to the nature of the corpus and the language. However, the situation changes when we intend to integrate newer languages and ones that are dialectally spread. In this case we try to find the golden mean. On the one hand the primary corpus is preferably composed of the latest lexicon and on the other logical reasonability is expected with regard to the display of dialects. In the case of Nenets we include two languages (Tundra and Forest), in the case of the Enets too (Tundra and Forest). Sometimes we  should process language data whose dialectal category/classification is uncertain. In such cases the relevant language is the default one.

4.4.2 Diachrony and synchrony

Hopefully, it can be seen in the dictionary that, despite our intentions, the synchronic and diachronic approach blend. Thousands of papers deal with the question where the border between semantic change and polysemy is (e.g. one of the latest is Zalizniak 2008). In the case of an underdocumented language it can be much more problematic, since in many cases it is hard to differentiate a semantic change from a polysemic pair – lack of sufficient and relevant data. Leastwise we always have to take stand on the problametic or uncertain relations – we have some tools for marking the questionable cases: e.g. marking the certainity of the labelling of the classification, or giving an explanation in a commentary section.

5. Further aims for development

In the course of the development of the dictionary we intend to develop a dictionary that is flexible with the possibility of enlargement. In this section I present two of these plans.

5.1. Areal dimensions

The dictionary will visualize not only the ancient (Proto-Uralic and Proto-Samoyed origin) part of the Nganasan vocabulary, but it will contain different kinds of borrowings. As a result, when we enter build a loanword into the system, we start building the section of that language: e.g. Evenki, Ket, Nenets, Enets or Russian.

5.2. Genetic dimensions

During the development it is important to pay attention to the oldest stratum of Nganasan and thus consider the historical-etymological literature. The history of Nganasan is not possible to divide different periods basing on the sources Tthe only exception is Matthias Alexander Castrén’s material from the 19th century, although a quantity of this material is hardly sufficient for deep lexicological analysis). As follows we can correlate the Nganasan lexemes mainly to the proto-languages. We handle the proto-languages as natural languages in the dictionary due to practical reasons. Differences between duaghter languages and the proto-languages only occur when qualifying the formal and cognitive relations. 

Proto-Uralic (UEW, Normanskaya – Dybo 2010)

>>> Proto-Samoyed (SW, Helimski 1997 etc.)

>>> >>> Proto North-Samoyed (Helimski – Anikin 2007, Janhunen 1975, Szeverényi 2008 etc.)

>>> >>> >>> Nganasan 

References

Aikio, Ante 2002: New and Old Samoyed Etymologies. Finnisch-Ugrische Forschungen 57. 9–57.

Allan, Kathryn 2007: Metaphor and Metonymy. A Diachronic Approach. Publications of the Philological Society 42. Oxford: Wiley-Blackwell 

Anikin, A. E. [Аникин, А. Е.] 2000: Этимологический словарь русских диалектов Сибири: Заимствованния из уральских, алтайских и палеоазиатских языков. Москва–Новосибирск: Наука

Blank, Andreas 2001: Words and Concepts in Time: towards Diachronic Cognitive Onomasiology, <www.metaphorik.de 01/2001>[30 December 2012]

Fortescue, Michael 2005: Comparative Chukotko-Kamchatkan Dictionary. Trends in Linguistics. Documentation. Berlin – New York: Mouton de Gruyter 

Fortescue, Michael – Jacobson, S – Kaplan, L 1994, 20102: Comparative Eskimo Dictionary . Fairbanks: Alaska Native Language Press 

Fortescue, Michael 2011: The relationship of Nivkh to Chukotko-Kamchatkan revisited. Lingua 121. 1359–1376.

Geeraerts, Dirk 1997: Diachronic Prototype Semantics. Oxford: Clarendon Press

Geeraerts, Dirk 2010: Theories on Lexical Semantics. Oxford: Oxford University Press

Gévaudan, Paul – Wiebel, Dirk 2004: Dynamic lexicographic data modelling – a dictionary development report. In Proceedings of the LREC-conference, Lisbon 2004, Lissabon. Preprint: http://homepages.uni-tuebingen.de/paul.gevaudan/lrec_gevaudanwiebel2004.pdf  [30 December 2012]

Gévaudan, Paul 2007: Typologie des lexikalischen Wandels. Tübingen: Stauffenburg 

Grzega, Joachim 2002: Some aspects of modern diachronic onomasiology. Linguistics 44, 1021–1045.

Grzega, Joachim 2004: Bezeichnungswandel: Wie, Warum, Wozu? Ein Beitrag zur englischen und allgemeinen Onomasiologie. Heidelberg: Winter.

Helimski, Eugen 1997: Die matorische Sprache. SUA 41. Szeged: JATE

Helimskij, Jevgenij A. [Хелимский, Е. А.] 1986: Etymologica 1–48. Материалы по этимологии маторско-тайгийско-карагасского языка. Nyelvtudományi Közlemények 88, 119–143.

Helimskij, Jevgenij A. [Хелимский, Е. А] 1992–1993: Etymologica 49–79. Материалы по этимологии маторско-тайгийско-карагасского языка. Nyelvtudományi Közlemények 93, 101–123.

Janhunen, Juha 1975–76: Adalékok az északi-szamojéd hangtörténethez: vokalizmus. Az első szótagi magánhangzók. Néprajz és Nyelvtudomány 19-20, 165–188. 

Janhunen, Juha 1977: Samojedischer Wortschatz = Castrenianumin toimitteita 17, Helsinki

Koch, Peter 2001: Lexical typology from a cognitive and linguistic point of view. In Haspelmath, Martin – Ekkehard König –Wulf Oesterreicher – Wolfgang Raible (Hrsg.): Linguistic Typology and Language Universals = Handbook of Linguistics and Communication Science 20/2, Berlin: Mouton de Gruyter, 1142–1176.

Koch, Peter 2008: Cognitive onomasiology and lexical change. Around the eye. In: Vanhove, Martine (ed.), From Polysemy to Semantic Change = Studies in Language Companion Series 106. Amsterdam/Philadelphia: John Benjamins. 107–138.

Koch, Peter – Marzo, Daniela 2007: A two-dimensional approach to the study of motivation in lexical typology and its first application to French high-frequency vocabulary. Studies in Language 31:2, 259–291.

Kosterkina, N. T. – Momde, A. Č. – Ždanova, T. Ju. [Костеркина, Н. Т. – А. Ч. Момде – Т. Ю. Жданова] 2001: Словарь нганасанского-русский и русско-нганасанский, Филиал издательство «Просвещение», Санкт-Петербург 

Kövecses Zoltán 2005: A metafora, Budapest, Typotex 

Lakoff, George – Mark Johnson 2003 [1980]: Metaphors We Live By. Chicago: University of Chicago Press

Nikolaeva, Irina 2006: A Historical Dictionary of Yukaghir. Trends in Linguistics. Documentation. Berlin – New York: Mouton de Gruyter 

Normanskaya, Julia V. – Dybo, Anna V. [Нoрманская, Юлия B. – Дыбо, Aннa B.] 2010: Тезаурус. Лексика природного окружения в уральских языкaх. Москва: Тезаурус 

Szeverényi, Sándor 2008: Tulajdonságfogalmak lexikai kategorizációja a nganaszanban. [Lexical Categorization of Property Concept Words in Nganasan] PhD dissertation. Szeged.

Tucker, Gordon H. 1998: The Lexicogrammar of Adjectives, London – New York, Cassell

Veenker, Wolfgang 1975: Materialien zu einem onomasiologisch-semasiologischen vergleichenden Wörterbuch der uralischen Sprachen = Hamburger Uralistische Forschungen I, Hamburg

Wagner-Nagy 2006: Nganaszan morfológiai szótár. Ms.

Wagner-Nagy Beáta é.n.: PS etimológiák és nganaszan megfelelőik. Ms.

Werner, Heinrich 2002: Vergleichendes Wörterbuch der Jenissej-Sprachen 1-3, Wiesbaden: Harrassowitz Verlag 

UEW = Rédei Károly 1986: Uralisches Etimologisches Wörterbuch, Budapest: Akadémiai Kiadó

Zalizniak, Anna 2008: A catalogue of semantic shifts: Towards a typology of semantic derivation. In Vanhove, Martine (ed.), From Polysemy to Semantic Change: Towards a typology of lexical semantic associations. Studies in Language Companion Series 106. Amsterdam/Philadelphia: Benjamins. 217–232.