The last 25 years of toponymy in Hungary in a way can be characterised as contradictory. In certain fields-study of macrotoponyms (especially of the names of settlements), or theoretical onomastics-outstanding issues and remarkable synthetic works have been born during these years. The research on microtoponyms can also be characterised as rich in detail as well as many-sided. Nevertheless, in this field no synthetic work has been published to sum up the whole system of microtoponyms. Researchers have been mainly engaged in the work of collecting place names, which revived in the mid-1960s. As a result of this work, today there are toponymic data available in the hundred thousands in collections of high quality, done with a more or less homogeneous approach.
If we want to have a homogeneous view of the system of Hungarian toponyms -its general interconnections, its differences and similarities in the course of history, the basic forms of its changes, and the directions of its constant augmentation-we need something which is indispensable: a homogeneous descriptive scheme, and the consequent net of notions which satisfy the demands of an up-to-date “toponymic taxonomy”. My objective in this work has been to make an attempt to create an analytic model of this kind.
It has always been important from the beginning of Hungarian toponymic research to try to depict the whole system. In the last century, when the first steps were taken, this idea was merely articulated as a demand for a systematic collection of names. Later, in the first three decades of our century, it was the etymological analysis of names that achieved paramountcy among the several modern analytical points of view in onomastics, which is quite understandable if we consider that it was the historical aspect that dominated the research of the time. This type of survey lead to great synthetic achievements in the oeuvres of two linguists: JÁNOS MELICH (A honfoglaláskori Magyarország. Bp., 1925-9) and ISTVÁN KNIEZSA (Magyarország népei a XI. században: SztIstván-Eml. II., 365-472; Keletmagyarország helynevei: MagyRom. I, 111-313). The differentiation of the etymological layers of our toponyms-especially macrotoponyms-, and their grouping on a chronological basis, served as important reference points, mainly in historical research.
Kniezsa’s historical typological system of toponyms was further refined by GÉZA BÁRCZI (A magyar szókincs eredete. Bp. 1958. 145-51), and lately GYULA KRISTÓ has made an attempt to dissolve the inherent contradictions mainly characterising the chronology (Szempontok korai helyneveink történeti tipológiájához. Szeged, 1976.). The system of Hungarian toponyms as a whole has recently been surveyed by BÉLA KÁLMÁN (A nevek világa. 1st publication: Bp., 1967, 4th publication: Debrecen, 1989; English version: The World of Names. Bp., 1978). LAJOS KISS’s work, “Földrajzi nevek etimológiai szótára” (1st publication: Bp., 1978; 4th, enlarged and revised edition in two volumes: Bp., 1988), is a large-scale summary of Hungarian toponyms. His work is also remarkable from the point of view of toponymical taxonomy. The author’s homogeneous categories and consequent approach to name analysis can be clearly seen in the background of the etymological explanation of placenames.
A special interest in microtoponyms has characterised Hungarian linguistics since the 1930s. The change in attitude towards toponyms could be seen first in the works written by the representatives of the so called Transylvanian school”, especially in those of ATTILA SZABÓ T. The first works to give an appropriate scheme for the analysis of this layer of toponyms were LORÁND BENKŐ’s “A Nyárádmente földrajzinevei”. MNyTK. 74. sz. Bp., 1947, and LAJOS LŐRINCZE’s “Földrajzineveink élete”. Bp., 1947. These works were later followed by two others-having similar objectives-written by MIKLÓS KÁZMÉR (Alsó-Szigetköz földrajzinevei. MNyTK. 95. Bp., 1957) and GÉZA INCZEFI (Földrajzi nevek névtudományi vizsgálata. Bp., 1970.). The main aspect in the analysis of microtoponyms-within the boundaries of onomastics, which was gradually gaining explicit independence-was the so-called “toponymic physiology”. To a certain extent these works bear the marks of the psychological trends in linguistics of the first half of our century.
Nevertheless, to weaken the one-track approach, researchers tried to take into consideration not only the different situations of assigning of names, but also the linguistic structure of names, as well as the types of toponyms. Nevertheless, these aspects were not united under the umbrella of a homogeneous theory. The potentially justifiable levels of analysis were mixed with each other, so the categories defined on a heterogeneous theoretical basis were mixed up, too. Because of this-and not their linguistic transience or synthetic character-certain toponyms can be classified in more than one typological group at a time.
The framework of these toponymic typologies has gradually been demolished in the recent decades by the new results in this field. The growing stock of toponymic data encouraged partial research work to flourish. Nevertheless, it has been a comparatively slow process for the achievements in general toponymic theory to penetrate Hungarian toponymic research, despite the works of ANDRÁS MARTINKÓ and JÁNOS BALÁZS on the one hand-published as early as the 50s and 60s-which already dealt with very important questions of this field, and that of KATALIN J. SOLTÉSZ on the other hand (A tulajdonnév funkciója és jelentése. Bp., 1979.), which was an application of the most up-to-date achievements of universal linguistics and onomastics to the Hungarian language. What the author says about the general characteristic features of proper names as linguistic signs (concerning form and meaning) by all means has to be considered in every particular field of onomastics, thus also in microtoponymic research.
Certainly, in the attempt to construct a model for the analysis of toponyms it is not enough to take into consideration what is said in this field within the national boundaries, but also what has been achieved in the field of microtoponymic research outside Hungary. Since the beginning of the 1970s it is analogy, and the question of toponymic systems and models that has come to the forefront of toponymic research in Europe. This direction has been shown by the fact that although the researchers had different reference points-some started from the general level of the theory of communication (ŠRÁMEK, Zum Begriff “Modell” und “System” in der Toponomastik: Onoma 1972/73: 55–75), while others undertook to describe a huge number of concrete toponyms (KIVINIEMI, Perustietoa paikkanimistä. Helsinki, 1990)-their consequences are very similar to each other. A system of toponyms can only be described by exploring the toponymic norms and the “naming models” existing in the linguistic consciousness within a language community. The rules governing naming do not only characterise a toponymic system at a certain point on the time axis, but also have an effect on the characteristic features of the newly born elements of the name system-resulting from the hereditary nature of the model effect.
It is more appropriate to construct an analytic model for toponyms on the basis of theoretical considerations; the analysis of a concrete collection of names, of a closed corpus, is less suitable when one aims at constructing a toponymic typology that can also be used in the analysis of a great number of toponyms. In this work I also had to undertake to reconsider and renew the terms and categories used in toponymic description, and-so to say with transposing the point of view-to integrate them in a new network of relationships, with the aim of enforcing new, different approaches.
Despite the theoretically based approach my analytic results are always illustrated by real toponyms, thus checking theory against practical reality. In certain fields (eg. semantic naming, or constructional changes)-lacking precedents-I also had to do basic research work, thoroughly examining certain layers of concrete toponymic systems.
My approach in this work is typically microtoponymic. This fact can be explained by several causes. On the one hand, linguistically the course of creating and modifying names is much more blatant in this youngest layer of toponyms than in other groups of names. On the other hand, microtoponyms form the widest layer of toponyms in number-and consequently also in diversity.
In constructing the analytic model of toponyms I have relied on linguistic principles, exclusively applying the methods of this field of research. “Name” is a linguistic phenomenon, and as one of the specific features of humanity, it can be the subject of several disciplines and examinations. Toponyms can serve as additional material for many branches of research, and we also have to adapt the results of other fields of research in their analysis. Although we do not question the legitimacy of a complex analysis of names, in certain respects we may consider it appropriate to narrow the horizon of research; concentrating on one sole aspect, through homogeneity we can explore such phenomena that can serve as basis for comparison for other, different approaches.
With the help of the methods of linguistics, and its complex analytical methods toponym researchers have proved that the structural, semantic and stylistic questions of names, as well as the examination of their social use, etc. can on their own serve as fully efficient analytical bases. If we want to describe the systemic character of toponyms, their inherent relationships, or the operation and changes of the system, we have to define an analytic frame that seems to be most characteristic of both the toponyms’ onomastic nature and generally their being linguistic signs. The results of specific onomastic and general linguistic research unanimously prove that it is structural analysis with a functional aspect that meets these demands to the greatest extent. Within the frames of an analytic model like this we have to show the common features of the category of toponyms together with the internal complexity of the system.
While retaining the homogeneity of the linguistic analysis in the description of toponyms we have to separate clearly the two kinds of analysis-which for quite a time in the history of linguistics have been distinguished: the descriptive and the historical. However, this separation is to be made with respect to the set of notions of both analytical methods: they have to be compatible, thus making it possible to show functioning and changing of toponyms as phenomena that are interdependent and mutually determined by each other.
The traditional linguistic and onomastic view is that-under the circumstances of natural, or popular naming-toponyms (first of all microtoponyms) can be described as proper names that had been born as common words defining places, which later gained their secondary status of proper names. This secondariness of proper names behind common nouns can be applied to the category as a whole, but it is not necessarily applicable to the individual names. What is true of a whole class is not necessarily true of the class constituents. Nevertheless, in every naming situation this secondariness appears: only a denotatum that is primarily denoted by a common noun can get a name. However, these common and proper names can be linguistically entirely independent from each other.
Proper names are a linguistic universal, ie. there are no languages lacking proper names. Thus the creation of toponyms has to be regarded as a phenomenon under the umbrella of a current system of proper names. The base of naming is a linguistic competence. If you want to name something, the regular (and only possible) way to do it-according to this competence, which is necessarily based on reflecting certain regularities-is to create a linguistic product, the place of which in the system had already been prepared, ie. this linguistic product will serve as a proper name for others as well as for the ‘inventor’. So, naming at any time is based on obeying the rules of an existing onomastic system.
Resulting from what is said above we can see that in naming instinctiveness and consciousness cannot be confronted as concepts excluding each other: to a certain extent consciousness plays a part in all forms of naming. Motivation is another feature that characterises the act of naming. Nevertheless, the level of motivation is very diverse if we compare the individual toponyms in this respect. The minimum of motivation is the taxonomic determination, but in most cases the denotata themselves prescribe the toponyms, in the sense that their inherent characteristic features-being reflected by man-serve as a base for their naming. The new name cannot contradict objective reality. Thus a name at its birth is necessarily descriptive and semantically always conscious. Though, theoretically it is possible that a person, while naming a place, coins an entirely new linguistic unit, this would contradict linguistic economy-a general principle, also governing the category of proper names. That is why the units of the new toponyms are derived from the existing set of linguistic elements.
Getting acquainted with and using the stock of toponyms, individuals also learn certain name-forming rules-as happens with common names. So, it is only com- munication that can form the speaker’s specific onomastic consciousness and can determine their onomastic competence. On the one hand, this ability of the speakers ensures them that they are able to recognise certain-even unknown-linguistic formations of speech as toponyms, or on the other hand it makes them able to create such formations of signs. The principal understanding that the models of the existing toponyms determine the types of the nascent names and also directly the names themselves, makes it possible to distinguish the two levels of linguistic description: the descriptive and the historical approaches. If we describe the use of names as a linguistic norm-by exploring certain regularities-we can comparatively easily approach from here to the questions of how names are formed and how they change, and the other way round as well: regularities in changes draw our attention to the rearrangement of the inherent relationships in the system. Thus separating the descriptive and historical analyses exactly serves the purpose to make clear their relationship, ie. to see more clearly how the system works and explore its forms of motion.
The natural basic unit of toponymic research is a settlement: mostly it is the inhabitants of a village, or a smaller town that form an ’onomastic community’ within which new names are created in a way that is determined by the system, too, and directed by communicative needs.
The onomastic formations of the community members can only be accepted by a social environment that has a roughly similar knowledge about the surrounding material and intellectual world, an environment that shows only a minimal difference in language and toponymic knowledge when compared with the other.
Within the individual ’onomastic communities’ (in a wider sense: within dialects, or languages) from age to age there are regularities characterising the system and the formation of names. The set of rules that play a part in the functioning of toponyms and in their formation can be referred to as the toponymic norm of a language. Resulting from the systemic nature of name formation, the toponymic system of the individual language communities-within areas showing linguistic unity, historical and cultural similarity, and minor differences geographically-does not seem to be substantially different. On the contrary, when we examine a suitable amount of toponyms collected from a large area-resulting from the differences in the above-mentioned factors-it is rather the differences that call our attention. Defining the borderlines of different phenomena in the toponymic norm of the parts of a linguistic area, we may be able to try to describe the Hungarian toponymic geography, ie. the ’dialects’ of toponyms.
The most important characteristic features of the naming norm of a given language territory can be most easily explored by the descriptive structural analysis of toponyms. In a structural analysis names can be examined in two ways: from a functional-semantic, and a lexical-morphological point of view.
The functional-semantic analysis enumerates all the model types that may serve as bases of approach in naming. The resulting categories are not linguistic classes, but epistemological categories of human thought. In them the characteristic features of the denotata-recognised and reflected in the given names-are made explicit. Thus, they can only be characteristic of a language in the sense that they show differences in the methods of naming: if we examine which forms of naming, which are available in a relatively great number, are preferred more or less, or even abandoned in a given era by the members of a certain language community.
Functional-semantic analysis means the semantic and linguistic identification of naming motives. The basic concept of the analysis is the name constituent: a unit of the toponym, which-in the situation of name formation-expresses any semantic feature that is connected with the signalled denotatum. So, in the course of analysis we examine the linguistic elements of a name with the aspect of their relationship with the denotative meaning, which is the semantic core of toponyms. As we can see, ’name constituent’ is a concept expressing relationship. That is why two formally identical sets of elements do not necessarily function as name constituents of identical value in different names.
Name analysis and the functional-semantic classes themselves seem to be closely related to the individual types of denotata. Our analytic model should be sensitive to this dependence. In order to meet this demand-as a preliminary-we have to categorise the types of toponyms in a way that reflects the differences of the classes of denotata. From the point of view of toponyms it is not the question of in what logical order we can describe the types of toponyms that is substantial, but it is much more important to show how this system is reflected in proper names. In the classification of kinds of places denoted by proper names, the linguistic signs used to express them, the geographical common names may serve as the base. In my recommendation there are 7 main classes, each containing 3 to 9 sub-categories.
In order to be logical, we need to have a homogeneous set of concepts in the structural analysis of names, and for that we also have to consider the possible functions of name constituents. Name constituents in toponyms basically can have four different functions: 1. they designate the kind of place; 2. they denote the place itself; 3. they express one of the features of the place, and in certain names 4. the expressed function is not related to the denotatum (within this category-the ’reminding’ function: Rákóczi/utca [Rákóczi street], and the ’conventional’ function: Fecske/utca [Swallow street]). The further differentiation within the designation of the types of places means that we have to differentiate names according to the common geographical nouns they contain: Bakony/ér [Bakony/brook], Kis/hegy [Low/mountain]. In denotative function one of the two segments of the name is always another authentic toponym. The name constituents’ functions expressing a feature of the place semantically have many variations. There are three main categories to show this variety: the name segment may refer to the attribute of the place (its size: Nagy/domb [Big/hill], its shape: Rövid/dűlő [Short/(larger)patch], its material: Homok/dűlő [Sand/(larger)patch], its colour: Vörös/föld [Red/ground], its age: Új/sor [New/row] etc.), the relationship of the place with something not in- herent in it (designating the flora: Hárs/hegy [Mount/Linden], the fauna: Ürge/domb [Gopher/hill], a building: Kórház/utca [Hospital/street], possession: Barta/tag [Barta/patch, ie. the patch is owned by a person called Barta] etc.), or the relationship of the place with another place (designating that the place is part of something: Hegy/hát [Mountain/back], designating the precise or relative position of the place: Sédi/dűlő [’a (larger) patch on the place called Séd’], Alsó/rét [Lower/field], designating direction: Esztári/út [’a road leading to Esztár’].
Realisation of these functions in names may serve as base for a structural analysis. Examining the functional structure of names we can see that the majority of our toponyms are single-, or double-constituent names. Names with three constituents are very rare in the Hungarian toponymic system. The basic set of our name system is unquestionably formed by two-constituent names (especially the variety expressing attribute + kind of place: Öreg/utca [Old/street], Büdös/kút [Stinking/well]). When I refer to the parts of the double-constituent names I use the terms ’compliment constituent’ and ’basic constituent’. If we want to distinguish the types of double-constituent names on a structural basis, we have to enumerate all the real possibilities of linking the name constituents in this type of name.
Single-constituent names can be characterised not so much by their functional features, but by the tendencies and characteristics in their formation.
Another possible level in the structural analysis of toponyms is the lexical-morphological analysis. In the course of this analysis names, and name constituents are examined with the aspect of the linguistic means of expression used in them. When we describe the toponymic system on this level, we are examining the possibilities how the functional-semantic categories of name formation are realised, by what lexical and morphological means. The basic concept of the lexical-morphological analysis is name element; an umbrella term for all the lexemes and suffixive morphemes (derivational and inflectional suffixes) that take part in forming the name. In order to describe the lexemes forming names and name constituents we should enumerate their characteristic features in the categories of parts of speech, and semantics: 1. common name (common geographical name, national name, name of occupations, name of plants, name of animals, name of materials, name of buildings etc.), 2. place name or its derivative (single-constituent, double constituent, derived, non-derived names, element of a place name), 3. personal name (family name, Christian name etc.), 4. other proper name, 5. a word with an adjectival character (derived, non-derived, participle etc.), 6. numeral (cardinal or ordinal), 7. syntagm (adverbial; verbal, participial; inorganic).
There is no hierarchical dependence between the functional-semantic and lexical-morphological model, they are just placed on different levels. It is true for all languages that there are many formal possibilities available for the actual realisation of the functional categories. The cause of this variety is that the lexical-morphological models of names are attached to the morphology and syntax of the common nouns.
In order to describe the syntactic relationship between the constituents of names containing two constituents we put them into categories according to the syntactic types used in forming these names: place names can be adjectival, adverbial and co-ordinative constructions. From the descriptive point of view it is the adjectival group that mainly causes problems, since the terms used to describe them in the grammar of common nouns (qualifying, specifying etc. premodifiers) are less suitable here. The complexity of this name type can be de scribed more precisely with the notional concepts of our semantic model. To this we can at most attach the phenomenon of not signalling the adjective constructions (Görbe/ér [Crooked/brook]), or signalling them (Falu/erdeje [’The Village’s Woods’]).
Describing the ways of linking these functional-semantic and lexical-morphological structure types that are realised in names, we can present important features of the examined ’name community’s’ naming norms in the field of toponyms. It is not without problems if we want to co-ordinate the functional-semantic categories and the lexical-morphological models: on the one hand, the functions of the name constituents are not at all subject to rapid changes, if they change, it is just their variations that change, on the other hand, the lexical-morphological models, which are linked to them, show a relatively rapid change-as linguistic signs usually do. Thus from time to time there is a kind of tension arising in the naming norm between the two levels of the name structure: new linguistic structures are attached to the existing functional models.
In the analysis of the history of name formation we have to take this above-mentioned change into consideration. In the course of the historical toponymical analysis we are examining what linguistic rules govern the formation of new toponyms and what kind of forces rule over the linguistic units when they are built into toponyms.
When describing this, our basic principle is that existing toponyms-through their model effect, by means of the speakers’ name competence-fundamentally determine also the character of the new elements of the system. Nevertheless, the constant changing of the names’ lexical-morphological character is not exclusively an aspect of the formation of new names, but also plays a part in the modification of the existing toponyms.
The possibilities of toponym formation can be described in the following five categories:
1. Names formed by syntactic construction are originally created as syntagms. Both units of the construction give a piece of information about the denotatum, so they are to be regarded as name constituents expressing a functional feature. The most important type here is that of the names based on attributive phrases, which at the same time is the central type of the Hungarian toponymic system. In synchronic systems it is not easy to separate the names based on adverbial phrases. The problem is that names formed in a different way also contain adverbial syntagms, or adverbial suffixes. A similarly special layer of names is that of names based on co-ordinative structures. These latter-with only some exceptions-are artificially formed names, mainly induced by administrative amalgamation of settlements (Budapest, Baktalórántháza).
2. The term morphematic construction covers a kind of name formation in which a bound morpheme (derivative or inflectional suffix) or a functionally similar element (postposition) is added to a lexeme, thus enabling the lexeme to function as a toponym. So, name formation can be defined as a process, during which the toponymic, ’proper name’ status of a given string of signs is created by adding a toponymic suffix to a stem morpheme: Bogáncsos, Kereki. Sometimes in practical name analysis it is very difficult to distinguish derived names from names containing derivational suffixes, since the latter may also be formed by grammatical construction, structural change etc.
From the group of modificational nominal suffixes it is only the plurals and the possessives that function in name formation. The group of toponyms formed with the plural suffix (Nyilak, Nyolcasok) is much more complex semantically than the other, formed with the possessive suffix (Almásié), which can only express one functional feature. Formally both groups are almost entirely closed: names belonging to them are not likely to be subject to further derivation, and they are seldom used as parts of other names. Both name groups seem to be very strongly fixed to certain areas. Construction of toponyms with adverbial suffixes (Csárdánál) and postpositions (Falufölött) is only related to the former group in terms of the expressed function, while in other respects rather differences are dominant.
3. In semantic name forming existing lexemes (common nouns and proper names) get a new, toponymic meaning. According to the characteristics of the semantic change five main types can be distinguished.
Semantic split means that a geographical common name becomes a linguistic unit valued as a proper name: Séd [Brooklet], Akácos [Acacia Grove]. This name formation is dependent on the character of the denotata, thus related to the dimensions of the given microtoponymic system. Names formed this way are often based on compound words, and are sensitive to structural changes.
Extension or narrowing of meaning (Pest ’Budapest’, Amerika ’USA’) can be understood as a change within one class of objects. Its condition is that the old and the new names live parallel in a polysemantic relationship. This type of name formation directly effects the hierarchical order of names, and it also has a major effect in toponymic taxonomy.
Two from among the forms of metonymy described in semantics are of great importance in the formation of toponyms. In our typology we have to distinguish quite a few variations of metonymic name formation. The majority of them are based on the local relationship of places (eg. a river → the area on which it flows: Séd ’szántó’ [Brooklet ’field’]. Metaphorical name forming has been almost entirely restricted to the type of names defining shape (Fecskefarok ’dűlő’ [Swallow’s Tail ’(larger)patch’]). But beside these, the metaphorical use of the names themselves is also important (Szibéria ’dűlő’ [Siberia ’(larger)patch’]).
Inductive (secondary) name formation proves the autonomy of the onomastic system: the existing toponyms-especially those with a metaphorical character-can directly induce forming names that have no causative relationship with their denotata; thus their descriptive character is very restricted: Csizmás ’ilyen alakú dűlő’ [Booted ’a piece of land shaped like a boot’] → Kapcás/tag ’a Csizmáshoz közeli szántó’ [Foot-clothed/land ’a field near the area called ’Booted’].
Removing names can be observed among microtoponyms as well. The base of this is similarities discovered between two landscapes, the toponymic mastery of the community that has moved to a new settlement, as well as the reminding function of names.
4. In names formed by a structural change the modification of the form of the toponym takes place while keeping the denotative meaning untouched. Thus what appears is always a synonym of the existing name. The smallest change that can be recognised from a structural point of view is a difference of one morpheme (reduction or expansion). If the name is reduced with a functional name constituent, I call it ellipsis (Hosszú/dűlő >> Hosszú). If one element of a name constituent is omitted, I call it reduction (Csorgóér/hát >> Csorgó/hát). The expansion of the body of the name can happen by completion-when a new name constituent is added–(Bogács >> Bogácsi/dűlő), or by augmentation-when a new element is added (Nyolcas >> Nyolcasi).
I am discussing folk etimology (Föhenyes/árok [Sanded/ditch] >> Fényes/árok [Bright/ditch]), and de-etimologisation here, within structural changes, as the denotative meaning of our names formed this way remains unchanged. The modification of the form can be smaller here-not necessarily one morpheme-, but they are never regular, systemic sound changes.
5. Name adaptation enriches the toponymic stock from an external store. Since their problem is substantially different from that of the above-mentioned processes of name formation, I do not discuss them in this work.
My analytic model first of all aims at describing basic relationships, and grasping general characteristics. We should not forget that constructing an ideal typology that is able to fully describe the formation and the structure of all toponyms-not neglecting even their irregular characteristics-is not very likely to be born. Language-and so the toponymic system-has always been characterised by features and processes living side by side, collaterally or contradicting each other. Where the main active forces meet-as the result of the interference-plenty of varieties and individual features arise. Aiming at a synthetic description one can neglect these variations, while a sensitive analysis can show the general active forces behind the peculiarities.
Certainly, my model for the analysis of toponyms will only be proved when it is used for the systemic examination of a great number of toponyms-which examination was the real goal of constructing it. I assume that if one uses this kind of analysis in processing toponymic data-even some 10 thousand of them-the experiences will surely demand the refinement of the categories and concepts of the descriptive model. Nevertheless, this type of relativity and open-endedness is not just a natural concomitant of the processes in scientific researches, but also a necessary requirement.