THE REGION AS A CONCEPT: TRADITIONAL AND CONSTRUCTIVIST VIEW

The main goal of this article is to assess and compare the various understandings of the concept of the region. The aim is to characterize the concept of a region as well as how its meaning has changed through geographical history, to mention the most important personalities and how they understood the concept of region. The article presents two different ways of looking at a region: 1) the region in the sense of traditional regional geography; 2) the region in the new regional geography (region understood as a social construct). The article then compares the two approaches and outlines both their advantages and their disadvantages. The first section presents a brief overview of how the understanding of the concept of region developed. The following part focuses on development of the concept of region as a social construct, especially in the context of the development of new regional geography, cultural turn and new regionalism. Finally, the article emphasizes the essential complementarity of the two approaches and briefly proposes a more complex scheme of analysis of a region.


Introduction
"Geography changes as society changes." (Livingstone 1992: 347) According to its advocates, regional geography is the core and heart of geography, the highest art of a geographer and the reason for its existence, and the advocates call for "back to the basics" (Whittlesey 1954;Hart 1982;Watson 1983;Lewis 1985). According to spatial scientists, regional geography is not exact, it does not search for laws and regularities, and its lack of a theoretical and methodological framework excludes it from the portfolio of exact sciences (Fred K. Schaeffer [1953] and other advocates of geography as a spatial science). Due to its philosophic-paradigmatic background there are also different views of the conceptual shape of regional geography, its idiographic or idiographic-nomothetic character, its focus on explanation or understanding and so on.
Many authors have participated in the discussion about the character of regional geography (Hartshorne 1939;Whittlesey 1954;Hart 1982;Johnston, Hauer, Hoekveld 1990;Entrikin, Brunn 1989;Nir 1990;Wood 1999;Claval 2007). On one hand there are the ever-strengthening positions of the advocates of "traditional regional geography" who emphasize a "return to the basics", "heart of geography", "nature of geography" (Hartshorne 1939), and a return to Hartshorne (Entrikin, Brunn 1989). Many of them stress the uniqueness of locations: "Hartshorne is correct about the uniqueness of locations" (Bunge 1979: 173). Their arguments are also supported by an emphasis on the importance of local scale in postmodern geography (Duncan 1996). On the other hand, there is a new (reconstituted, transformed, reconstructed) regional geography (Gilbert 1988) which started the exactization process of regional geography. This has meant there is a visible shift of approach in regional geography, mostly a more significant orientation towards processes and contexts (Tomaney 2009).
The difference between traditional regional geography and new regional geography (social-constructivist approaches in regional geography) has kept increasing gradually (Paasi 2009). The division of regional geography into two different approaches brought about several discrepancies. As a consequence, it influenced regional geographical practice and the way a region was understood, i.e.: what is a region (a complex unit or a social construct); what isn't a region (the social-constructivist new regional geography does not take nature sufficiently into account); how to investigate a region (social-constructivist approaches emphasize that contexts and underlying processes are important, whereas traditional regional geography is rather a descriptive science). Traditional regional geography uses traditional methods (statistical analysis, fieldwork, regionalization etc.), whereas new regional geography uses qualitative and contextual methods. Traditional regional geography attempts to "see the region objectively", whereas new approaches see the region more subjectively. This has led to our decision to focus on the meaning of the concept of region, and to focus on how this meaning developed over time. We also wanted to provide a comparison of basic approaches.
In this article, we focus on the changes in understanding the concept of region as follows: its complexity; its synthetic character; its unity; the role of man, nature and society in the formation of region; the interconnections of its individual parts; uniqueness; dynamic vs. static character; region as a result of development vs. region as a process. The difference between the traditional concept of a region and a region as a social construct (as understood in new regional geography) is as follows: As well as the term region, we also use the terms landscape (landscape, according to Carl Ortwin Sauer can similarly be understood as a region) and place ("Even for many new regional geographers, the meanings of region and place are more or less similar or overlapping" ;Paasi 2009: 224). This article was written by two authors. One is a regional geographer and presents his view of a region from the point of view of traditional regional geography. The other is a social geographer and represents the approach of social constructivism in new regional geography.
The resulting article focuses on how the understanding of the concept of a region developed throughout the history of geography. The main goal is to assess the various understandings of the concept of the region and to present the most appropriate conceptual framework for a region and understanding it. The authors attempt to find answers to the (following) research questions: How has the meaning (understanding) of the concept of region developed over time? How was the concept of region perceived by significant geographers? What were the weaknesses and strengths of the main approaches? What are the main contributions of the main approaches? How can the positive aspects (those bringing some benefits) of both approaches be used when characterizing a region? In the conclusion, we offer a proposal for an analysis (and of synthesis) of a region, using the methodological contributions of both traditional and social-constructivist understandings of a region. The article takes the form of a discussion between the supporters of the traditional meaning vs. supporters of the region as a social construct (Hart 1982;Hartshorne 1939;Johnston, Hauer, Hoekveld 1990;Murphy 1991;Paasi 1986;Sauer 1925;Semian 2016;Whittlesey 1954 etc.). This provides an analysis of the concept of a region in individual approaches, as well as an analysis of the concept of the region by different geographers. The comparison of different approaches (traditional vs. new regional geography) is based on an analysis of the strengths (primarily) and weaknesses. The strengths and weaknesses of these approaches (to the region) relate to the theoretical-methodological area (complex vs partial understanding of a region; static vs dynamic region; description vs contextual and processual understanding) and to applications (used in particular in regional development).

The concept of a region in traditional regional geography
Throughout the history of geography, the region was, and is, its most important topic, its main concept, and its main object of study. However, the concept of the region was understood differently throughout the history: a region was understood as a pure intellectual construction (Hartshorne 1939); as a concept or method (Whittlesey 1954); as a system (Nir 1990); as a total and complex unit (Paul Vidal de la Blache); or it was understood in the sense that a region is no more than a sum of its components (Hartshorne 1939), etc. During the 20th century, geography gradually split into two disciplines: human and physical (e.g. Hartshorne 1939). This was due to the following: 1) the importance of nature in the process of formation and development of the region kept decreasing; and 2) differences between the methodologies of natural and social sciences. As a consequence a region is understood as a social construct.

Origins of modern geography
Carl Ritter is the father of modern regional geography. He is the originator of new scientific geography, which is based on an organic unity between man and nature (Martin 2005: 125). "Ritter's … regional geography is conceived as unity in diversity; not an inventory, but an attempt to understand the interconnections and interrelations that make the area a mutual (zusammenhängig) association" (Nir 1990: 34). For Ritter, the earth and its inhabitants are in a close relation; the human and physical worlds are inseparable (Cresswell 2013: 40). In 1859, Darwin published his work On the Origin of Species. The subsequent approaches -social Darwinism and environmental determinism -explain regional differences as a result of the geographical environment. Such an approach had a decisive influence on geography at the turn of the 19th and 20th centuries (e.g. the determinist approaches of Friedrich Ratzel and other German authors, Ellen Semple, Ellsworth Huntington were prevailing).

Paradigm of regional geography
Starting from Vidalian geography we can notice a shift in the understanding of nature: man as an actor is being emphasized, and nature is perceived as a "product of the culture". Paul Vidal de la Blache is known as a representative of possibilism. As opposed to determinism, possibilism understands nature to be the possibility for regional differentiation. Vidalian geography and the whole of French regional geography were holistic and complex (Géographie Universelle as well as excellent local studies and regional monographs). Vidalian region is holistic and descriptive unit, with strong personality (Archer 1993: 499). Regional differences and pays, however, occurred mainly due to genre de vie. Man and human group can never escape the restrictions of the milieu, the natural surroundings upon which they depend (Mercier 2009: 148). It is necessary to add that milieu is not only the natural environment; Vidal makes a distinction between milieu externe (physical, not only natural world) and milieu interne (values, habits, customs).
Carl Ortwin Sauer, an influential American geographer, laid stress on culture in the landscape genesis, and he is thus a follower of the possibilist Vidal de la Blache. Carl Ortwin Sauer, however, includes nature in his view of landscape (region): "geography is based on the reality of the union of physical and natural elements of the landscape" (Sauer 1925: 325). He emphasized the division of forms into natural and cultural. The first part of his formal morphology includes both the reconstruction and the understanding of the natural landscape (ibid, from p. 333). It consists of geognostic and climatic factors, which are expressed in part through vegetation. Natural factors transform the natural landscape over time into forms (climate, land, sea and coast, vegetation), while creating a natural landscape. The second part of the morphologic analysis includes an analysis of the cultural landscape. Carl Ortwin Sauer argued that culture is the main agent in shaping the cultural landscape: culture is the agent, the natural area is the medium, and the cultural landscape is the result (Sauer 1925: 321 and figure, p. 343).
Both Alfred Hettner and Richard Hartshorne influenced the character of geography from the 1930s. William M. Davis described the landscape as a result of processes (such as erosion cycle etc.). His approach significantly influenced Hartshorne (Harvey 2009: 22), who wrote The Nature of Geography (Hartshorne 1939). The Hettnerian-Hartshornian chorology studied areal differentiation, and explained it by causal connections between phenomena. Hartshorne's chorology understands the region as a unique area and as a mental construct. In his diagram (Hartshorne 1939: 147, fig. 1). Richard Hartshorne placed an emphasis on regional geography, which in the physical-geographical and human-geographical point of view has a strong character of unity; physical geography is an essential part of geography (Butzer 1989). For Hartshorne, the region is the central organizing concept in geography (Smith 1989: 103). Regions are unique because they are unique combinations of phenomena. Hartshorne's approach is not problem-oriented; he wrote: "the interest of the geographer is not in the phenomena themselves, their origins and processes, but in the relations which they have to other geographic features (i.e. features significant in areal differentiation)" (Hartshorne 1939: 425-426).

Regional concept from the 1950s
American geography: inventory and prospect gives a deep insight into the perspectives of American geography. The main area of interest in geography covers areal differentiation; geography focuses on interregional similarities and differences, interconnections and movements and on the order found in space (Whittlesey 1954: 21). The region in Whittlesey's sense is a kind of a formal region. The region is a tool used by the regional method. Regions can be single, multiple or total. Complex regions are called total regions, compages: "Such a region is an association of inter-related natural and societal features chosen from a still more complex totality because they are believed to be relevant to geographic study. " (Whittlesey 1954: 35-36). Geography as a spatial science continues to understand a region in such a formalized way (Whittlesey 1954). This approach was dominant in 1960s and is typical for emphasizing the formal side of a region -spatial pattern, interactions, regularities etc. Spatial science is based on the philosophy of neopositivism and places an emphasis on formulating regularities. As spatial science does not deal with unique regions and their specifities, in this article we provide an analysis only of the traditional region and the region as a social construct.
The development of geography was further influenced by its division into physical and human geography and by its further fragmentation. The ongoing process of the loss of unity was associated with developments in geography (environmental determinism → possibilism → probabilism; from 1980s postmodern and post-structuralist geographies). The emerging new regional geography and cultural turn in the 1980s changed the perception of the region into a region as a social construct. New regional geography turned regional geographers into systematic geographers (Wei 2006(Wei : 1397. The understanding of a region as a social construct (in Marxist approaches, a region is produced) is obvious and some authors characterize it as "social (cultural) determinism" (Graham 1999). Due to shift to social constructivism, several articles about the social construction of nature (Demeritt 2002;Evans 2008) and the social construction of scale (Marston 2000) were published. However, Gerard A. Hoekveld identified a new conceptual framework for regional geography, including 8 key concepts, of which only the seventh is nature, with a note: "In regional geography nowadays [nature] is still conceived in a more limited way. " (Hoekveld 1990: 27) And finally, the "more traditional" Israeli regional geographer Dov Nir wrote: "Regional geography deals with the challenges posed to a certain society at a certain place on the globe and with the responses made by that society. Its focus is the study of differentiation between societies …" (Nir 1990: 2). A divided geography, however, is "weaker", its competitiveness and reputation fades (e.g. Matthews, Herbert 2004;Castree, Rogers, Sherman 2005).

The region as a social construct
3.1 From traditional to new regional geography Regional geography primarily studies the relationship between humans and the environment they inhabit. "Traditional" regional geography encompasses distinct notions of that relationship, i.e. what is the character of the relationship between man and his environment (deterministic -seen from one direction [Ratzel, Semple] or from the opposite one [Durkheim], and possibilistic [Vidal de la Blache]). Regional geography distinguishes geography from the other "big" fields of science because it is interested "in everything", although within a specific region, or, more precisely, because it studies and explains the differentiation between territories (regions). From the point of view of practical applicability, it abounds with great potential (regionalization, regional development).
Regions are not a purely geographical domain but are also used in many other fields -either as a method (a methodological approach to regionalization, e.g. comparing various regions in geopolitics), or as a tool/ purpose (a pragmatic approach to regionalization -creating regions in order to establish, for example, electoral districts). Regionalization as a method has in fact "endured" even the harsh criticism of regional geography by so-called 'spatial science' in the period following the Second World War.
During the next paradigmatic turn and in the face of criticism from spatial science (which is unable to explain the differentiation of regional development, the way a particular regional organization was established, and the like) the dualistic concept gained strength in geography, which was by then splitting into human geography and physical geography. Regional geography, however, has the advantage of being able to work with knowledge from both these geographical disciplines which sometimes are separated in an overly artificial, dichotomous manner. That is where we see a great "strength" of regional geography.
From the 1960s, roughly, and then during the 1970s and 1980s -after positivistic spatial science encountered sharp criticism for its detachment from the reality of social and political affairs, the "dehumanization" of human geography -human-geographical paradigms have been fragmented into many various directions responding to the diverse problems faced by society: radical geography, humanistic geography, feminist geography, etc.
Regional geography was not the only field to have undergone this change in thought, focused on the influence of culture and society, as it also occurred in other branches of social and human sciences and which is generally referred to as the "social" or "cultural turn" (Barnett 1998(Barnett , 2009). Another response of regional geography to the cultural turn is, besides the aforementioned multi-paradigmality, its multi-disciplinarity, i.e. adopting and applying methods and knowledge from other branches of social sciences and humanities (e.g. sociology, economy, psychology, historiography and many others).
In relation to regional geography (which has often been regarded as "dead", namely by the adherents of spatial science; Gregory 1978), humanistic geography in particular is understood as a "return" toward the idiographic approaches of traditional schools of regional geography. It is not only about a simple return toward an idiographic conception of space; even though humanistic geography is once more concerned with the uniqueness of specific places or regions but primarily from the perspective of the essence of such uniquenesses, from the perspective of subjective meanings that a person (both the one in the studied environment and the one studying a given environment) attributes to a particular place/region, influencing the given place/region by her/his perception -here we see one of the roots of the so-called new regional geography. In today's post-structuralist new regional geography, a region is perceived as a social construct continually endowed with subjective meaning and -just as in the case of an individual -characterized by a multi-layered identity (region as home, region as a political entity, region as an administrative unit, etc.).

The region as a social construct
Region specificity and incommutability had already been emphasised by Richard Hartshorne who claimed that a region was an arbitrarily delimitable territory, i.e. a sovereignly subjective matter. Even despite the prevailing systematization characterizing his approach (wherein chorology [regional geography] should involve "knowing everything" about a given territory [based, among others, on traditional German regional geography coined by Alfred Hettner]), his book titled The Nature of Geography: A Critical Survey of Current Thought in the Light of the Past (1939) can be understood, owing to the idea of the region as a social construct, as a "bridge" between traditional approaches to regional geography and the new regional geography. The concept of a region as a social construct later became the key concept within the new regional geography (Thrift 1983;Paasi 1986;Gilbert 1988;Murphy 1991;Schmitt-Egner 2002;Claval 2007).
Obviously, the concept of region as a social construct involves an enormous influence of culture and identity, or, more precisely, that of the cultural, historical and geographical context which plays a cardinal role in the formation of regions (other crucial concepts/key terms of both the new regional geography and the new cultural geography).
Traditional regional geography understood regions as a consequence of the interactions between society and its environment. From the point of view of the new regional geography, research into the interactions between man and nature, or rather, between society and environment, is being replaced with a study of the interrelationship between individual and society. The new regional geography no longer asks merely about "what", "where" and "when" but is interested primarily in the formation process of the region, in the way regions come to existence, for what reason and for what purpose they arise.
Various forms of the concept of a region can be found (not only) in geographical research. All the approaches mentioned above, regions created for the purpose of determining statistical or administrative units, "natural" regions as results of synthetizing analyses generated by traditional regional geographers are always, in a sense, a man-made construct. The region defined in this way is articulated from above by researchers, politicians and other actors; it is a secondary outcome of that particular activity in the course of which the region was established. This is the essential distinction from understanding a region as a social construct within the realm of the new regional geography. Here, regions are not approached as objects of study but rather as subjective constructs, a socio-spatial process. Regions arise from regional, social interactions that take place among individuals, groups and institutions in regional areas. Allen, Massey, Cochrane (1998: 50) suggest that a region is "the product of the networks, interactions, juxtapositions and articulations of the myriad of connections through which all social phenomena are lived out". Regions arise from interactions occurring at different hierarchical and scale levels of the society, i.e. through the actions among individuals, groups, institutions both within and outside a given region (Paasi 1986). As part of region formation, these relationships (all of them) are seen as reciprocal; constituting a condition for these interactions while being their result. It is not of substance whether or not a given individual considers a region to hold an important place in his/her everyday life, yet it is always produced and reproduced via ordinary activities. Kaj Zimmerbauer states that "at the core of social constructionism is the idea of region as a socially produced entity in which the regional consciousness of its inhabitants creates the whole idea" (Zimmerbauer 2011: 255). Individuals, groups and institutions active outside the region are of equal importance in the region-building process, regardless of whether or not they have the power to influence that process, and whether or not they do so deliberately (Paasi 2010).
Leaving aside all actors, their networks and mutual interactions, region formation is closely linked to the physical environment wherein a region is being constructed. A particular landscape and a specific natural environment markedly predetermines and affects both the material and the symbolic aspects of the region forming process, both its material form and its image (Šifta, Chromý 2014;Šifta, Chromý 2017). Many new regional geographers no longer pay much attention to the importance of the physical environment for regional formation. We do not suggest any return to Vidalian possibilism but 'pays' and 'genre de vie' cannot be entirely separated from the 'milieu' (the physical environment) wherein they get their shape and which they obviously influence and transform (Claval 2007;Paasi 2010).
This understanding of the concept of the region as a socio-spatial process has not, however, been unanimous. In the past three decades during which the region as social construct was establishing itself within the new regional geography (Thrift 1983;Pred 1984;Paasi 1986;Gilbert 1988;Murphy 1991), the understanding of this and the approach to it naturally differed in terms of both space and time. Following this initial stage of theoretical and conceptual development of the social constructivist approach toward the region, a wave of (neo-)regionalism could be observed in the 1990s.

Regions as a result of (neo-)regionalistic tendencies
In Europe, (neo-)regionalism manifested itself (in connection to the building of a 'Europe of the regions' within the EU) through an approach to regions from a political and economic perspective (Hettne 2005). We can distinguish two basic types of regionalism: one bottom-up and one top-down. The first one developed mainly owing to voluntary initiatives of citizens living in each particular region or those of local subjects (e.g. microregions, transborder Euroregions and consensual associations of municipalities, etc.). In the second type of regionalism, the development of regions is initiated (taking the example of Europe) by the EU's central institutions with the aim of enhancing regional competiveness and reducing socioeconomic gaps between the developed and the less developed regions (Bristow 2010). Thus new regionalists, by supporting not only socioeconomic, but also sociocultural development of regions (the forming of regional identity including its impact on regional development), respond to the deepening processes of globalization and unification (Chromý 2009;Paasi 2012). They emphasize and take as a basis regional diversity as well as the specific material and cultural values of the given region (Keating 1998;Chromý 2009;Paasi 2012;Jones, Paasi 2013).
Similar manifestations of (neo-)regionalism were also observed in the United States (e.g. Wheeler 2002) and in those Eastern European countries that are not (or were not) EU members (e.g. McMaster 2006).

The region as a brand
The economic or marketing concept of regions represents another approach to regions as social constructs, which was well-marked especially in the first decade of the new millennium. Place marketing and place branding researchers point to the fact that regions are treated as commodities in order to make profit (either by attracting investors, encouraging new inhabitants to move in, or by increasing the turnout of tourism). However, the majority of regions "operate" with place branding and place marketing strictly at the level of "selling" a region as merchandise, using its brands with the aim of commodifying and commercializing it without taking into consideration that the two concepts must be seen as a long-term strategic, synthetic and integral, complex process. This process, which makes part of an overall strategy of the given region for preserving and enhancing its competitive ability, is supposed to satisfy all target groups (Anholt 2003;Hospers 2011;Zimmerbauer 2011;Pike 2009Pike , 2011.

Overlapping regions
Another possible generalizing stream of working with the region as a social construct is constituted by "regional conflict" research projects, which we expect to grow in number in the near future. As is evident from the above, there are increasingly more regions of diverse character (administrative, economic and cultural; numerous tourist regions are emerging, NUTS system regions, transborder regions, all of them of various scale levels, etc.). Many of these more or less spatially delimited units overlap. Along with the changing context, many of them see their meaning change over time. Thus, conflicts of interest between different actors in regional initiatives become more frequent and regional identity becomes internally more fragmented. In addition, outward regional identity becomes ambiguous (Kašková, Chromý 2014).

The view presented by new regional geography
Owing to the revival of interest in regions within regional geography and beyond, research is becoming increasingly idiographic. When studying specific regions, new regional geographers, however, strive to reveal details on the functioning of regions, trying to make sense of the mechanisms of their formation, transformation and vanishing. Their objective is to interpret this idiographic knowledge, as it seems at first sight, by nomothetic means. The results of such efforts include, for example, Anssi Paasi's theory of institutionalization (Paasi 1986) as well as plentiful attempts to put this into practice (testing the region institutionalization process on specific regions). It is thus a combination of idiographic and nomothetic approaches.
We can, however, ask whether the existence of regional geography is legitimate and necessary. The pieces of knowledge that we learn about a region (as the main research topic) can be simply extracted from all the other systematic subdisciplines of geography, or from other scientific fields as a whole. For example, Gordon MacLeod and Martin Jones (2001) claim that priority is no longer given to only one discipline (regional geography), as regions are consistently studied in the whole field of geography. Regional geography is thus not necessary, but regions are what is needed in geography (MacLeod, Jones 2001). Regional geography can still be substituted by using regions as a delimitation of where other disciplines should be applied. The strength of regional geography, however, is in its complexity of synthetizing such pieces of knowledge, analysing them through a perceptive approach and allowing for the historic-geographical context of development in the studied region. Nobody but "complex" regional geographers can adopt such an approach which is crucial to not only understanding the formation process, existence and functioning of a region, but also to applying it, for example, in regional development.
When perceiving the region as a social construct, the strengths of such an approach include the following: -The nomothetic character of such an approach, which is achieved by providing an explanation of processes and contexts; this is a significant methodological contribution. Contexts and processes enable a better understanding of functioning of regions and thus predict their future changes. -Focus is given to those social topics, the significance of which within the region is growing constantly. -A greater emphasis is put on those concepts which were neglected in regional geography in the past: political power and the whole of politics; social differences and social changes; global and local scale etc. That enables a better understanding of the current state of a region. It is necessary to point out that weaknesses include, in particular, the following: -Nature is missing; there is a non-complex character; -Too much emphasis is given to social problems.
-Weaknesses relate to, in particular, the somehow reduced character of a region (the region is not so complex). -Solutions to problems in a particular region created and suggested within new regional geography research cannot be fully transferable to solutions of similar problems elsewhere (due to specific conditions and time-space context).

The view presented by traditional regional geography
Understanding a region as a social construct has some weaknesses. It is obvious that the importance of society is growing -and as a result the region as a social construct is becoming more and more important. Despite that, the role of nature cannot be ignored (global warming, natural hazards, etc.). Within regional differentiation, nature is still the real power. The division of Canada into heartland and hinterland cannot be explained only by communication connections and economic advances, as they themselves are a result of climatic conditions. A different understanding of the concept of the region is questionable. The increasing influence of reductionism in regional geography may be subject to criticism. Not only is the complexity of the region reduced, also the social component itself (the role of excluded minorities and different social communities is overvalued). Social sciences and geography still reflect social reality and now anticipate it, and they bring their own moral criteria to this. The identity of a region and region formation, as a theme, has been overestimated. Geography rejects tradition, it is "revolutionized". A positivist "epistemological turn" led to the formalization of the region; since the 1990s, an "ontological turn" has led to, it seems, growing vagueness and "mistiness" of geographical texts (see the increasing incomprehensibility of the fourth and fifth edition of The Dictionary of Human Geography). The traditional regional-geographical characteristics provide a more balanced, more complex, more usable (for planning etc.) and more vivid image of a region.
As to the traditional approach (the region in the sense of traditional geography), its strengths include the following: -a complex approach; well-balanced characteristics of individual spheres and topics; a systematic approach -focus is given to central (main) topics; -it is "demanded" by the public (a growing demand for regional information);. The weaknesses of traditional regional geography include the following: -its descriptive character; -little emphasis is given to society and to social topics; -static characteristics of the region.

Towards a more complex regional geography
The development of knowledge may be perceived as evolution, as a gradual addition of new ideas, contributions, methodologies, and procedures. Regional geography and the concept of the region may thus include contributions from spatial science as well as humanistic and radical geographies. They may also accept contributions from social constructivism. It is easier to understand a region when accepting humanistic-geographic concepts of topophilia, topophobia (Tuan 1974), the sense of place, and placelessness (Relph 1976); Marxists' concepts of social justice in the city and in rural areas (Harvey 1973); as well as the impacts of globalization and postmodern cultures on local environments and communities (Savage, Bagnall, Longhurst 2005), etc.
There are strong examples of "good regional geography" (more balanced and complex, more aimed at the most important phenomena) in the history of geography: Jordan's Texas emphasized the confluence of cultures (Jordan, Bean, Holmes 1984); Harm de Blij presented his deep understanding of the world by applying geographical concepts to world regions (de Blij, Muller 2010). Such regional geography can provide more complex studies of society, as well as studies focusing better on central problems and explanation. Dov Nir's conception of regional geography (Nir 1990) is based on systems theory. "Society and its physical environment is not a dichotomy: each is part of a whole, a system. " (Nir 1990: 8). Dov Nir introduces the concept of the region as a holon, "when viewed from the inside it is something closed, something final and defined, but when viewed from the outside appearing as part of something larger" (Nir 1990: 25). Dov Nir introduces the region as a system with phenomena that are components of a whole, with relationships between components, and relationships between components and their environment; system is more than the sum of its components. And Nir's model of a systemic region is a way to study "hidden factors" (ibid. p. 103). Instead of providing an exhaustive characterization of all the elements, a focus on the central issue is proposed (Nir 1990: 39;Baranskij 1953).
The authors present several proposals that are aimed towards better characteristics of regions: 1. Regions are complex and holistic in the sense of physical-geographical -human-geographical unity. 2. A region is an open system with its own structure and relations between its parts and components as well as relations between the region and its environment. 3. Emphasis should be laid not only on a detail description of the region, but also on the central issue and on the most important phenomena. 4. Regional analysis includes all the basic geographical spheres (natural, economic, cultural, social and political system); sub-spheres are not a must. Social sciences and new regional geography stress the importance of social factors and processes; social factors and processes (and relevant processes and actors) should be incorporated into regional-geographical research. 5. Characteristics of a region can be made "more exact" by including the processes, contexts and transformation, and by formulating research questions that would lead to explanation and understanding (Kasala 2014). 6. Regional geography must be more relevant, more practice-oriented, should fulfil public expectations and provide vivid descriptions. Regional-geographical characteristics may be identified by analyzing several "layers" gradually. Older approaches, which focus on the process of transformation, are of "Vidalian style" (i.e. they see the countryand-town symbiosis in the phases of historical succession [Wooldridge, East 1967: 158-159]) or they are in the form of Whittlesey's concept of sequent occupance. Sequent occupance of Southern California means the gradual transformation of the landscape in four stages: aboriginal -Spanish -American -international era. Niko Lipsanen's Master's thesis (Lipsanen 2001) offers three levels of analysis: the naturalistic analysis of Roseau (position, structure, function, texture); existential analysis (visiting, dwelling, changing); and synthesis (districts of Roseau, Roseau as a place). A triple model of place (Matlovič 2007) is composed of place as the filling of a part of timespace (physical and technical sphere components); as an arena, process -social construction (social sphere components); and as meaning, identity (noosphere and cyber sphere). John Agnew (2005: 89) presents an idea of three components of place: place as a location or a site; place as a locale (a setting for everyday activities); and place as a sense of a place (a place of identification).
"The ultimate goal of a regional descriptive synthesis was achieved through a thematic "layering" of subject matter, extending from the physical environment through several layers of human intervention. " (Pudup 1987: 1) In conclusion we would like to propose a scheme of layers of regional-geographical analysis. The analytical part of our research consists of three layers of analysis. The first layer is the "objective region". This layer provides an insight (detailed information) and broad understanding (comparative, processual and contextual). The second layer focuses on the personality of the region -by identifying its specifities, its central phenomena. And the third layer deals with subjective experience, sense of place, identity. Those three layers enable a synthesis and provide a deep understanding of the region. They can be a good basis for regional development and other applications.

Conclusion
"Regional geography cannot divorce itself from the empirical world. If it did, it would be likely to become a bloodless Platonic Universe of Ideas, merely producing theories for their own sake." (Wood 1999: 205) "The highest form of the geographer's art is producing good regional geography -evocative descriptions that facilitate an understanding and an appreciation of places, areas and regions." (Hart 1982: 2) One of the contributions of regional geography is that it defines regional differentiation and explains it. Changes within any scientific discipline are necessary; yet changes do not necessarily mean certain progress. Growth of knowledge is an evolutionary process. Our current knowledge is based on contributions which we "achieved" in previous periods. Traditional as well as new regional geography -both of them have advantages and disadvantages. Each of them can benefit from the other.
A comparison of these two basic approaches to the concept of region is one of the contributions of this article. By comparing the two approaches the authors present a brief proposal of a more complex approach in regional geography, showing that these two approaches are complementary, which is a benefit. Traditional regional geography is more complex, as it allows a better understanding of a region. On the other hand, new regional geography (as an example of socio-constructivist approach) is a contribution to geographical methodology, because it facilitates explanation by using contexts and processes.
In the introduction, the authors formulated several research questions. The 1st research question was: "How has the meaning (understanding) of the concept of the region developed over time?" The most typical changes in the meaning of the concept of the region included a loss of complexity as well as shift to a more social understanding. Regional geography gradually "split" into two main directions: traditional geography and new regional geography. The 2nd research question was: "How was the concept of region perceived by significant geographers?" The article focuses on key personalities -geographers and on their understanding of region. Carl Ritter looked for unity within diversity, interconnections and interrelations; for Ritter, human and physical worlds are inseparable. Starting with Paul Vidal de la Blache, nature is perceived as a "product of the culture". Vidalian French regional geography was holistic and complex. Carl Ortwin Sauer researched the landscape (i.e., region) by applying a morphological analysis which was composed of both analyses: analysis of the natural landscape and analysis of the cultural landscape. Richard Hartshorne understands the region as a unique area and as a mental construct. Hartshorne's chorology is typical of his strong character of unity.
The traditional understanding of the region underwent changes in its meaning in mid-twentieth century. Derwent Whittlesey (1954) understands a region as a formal region; and geography, as a spatial science, leaving the idea of a unique region completely behind, and investigating regional patterns, regularities, and interactions. Anssi Paasi and other representatives of the new regional geography understand the region as a social construct. The concept of the region presented by the Israeli geographer Dov Nir (1990) is based on systems theory. The authors identify the most important strengths and weaknesses (research question No. 3: What were the weaknesses and strengths of the main approaches?). The region as a social construct has advantages: a nomothetic approach with explanation based on processes and contexts; a strong emphasis on political and social themes; weaknesses (disadvantages) are the problem of transferability of solutions from one region to others; a non-complex character due to leaving out nature. The strengths of the traditional concept of a region are its complex and systematic approach; focus is given to central (main) topics, while the weaknesses of this traditional understanding of a region are its descriptive character and the static characteristics of the region with little emphasis placed on society. The 4 th research question was: "What are the main contributions of the main approaches?" The main advantages as and contributions of the traditional concept of a region include complexity and a focus on the main phenomena and specifities of region. The main contributions of new regional geography (the region understood as a social construct) include a greater emphasis given to social topics and methodological contributions (processes, contexts). The answer to the last research question ("How can the positive aspects of the two approaches be used when characterizing a region?") leads us to an attempt to find more optimal characteristics of a region.
As analyzed in the last part of the article, a more complex regional geography is based on systems theory (Nir 1990), a holistic complex understanding of a region (Nir 1990) as well as a socially produced and reproduced region (Gilbert 1988;Paasi 1986), and thus uses also new methodologies focused on processes and contexts (Johnston, Sidaway 2004). "Layering" of the research (see also Lipsanen 2001;Matlovič 2007;Agnew 2005;Pudup 1987Pudup , 1988) provides a deeper understanding of a region. The authors present "a model" with three layers of analysis, which include three ways of understanding (verstehen): 1) an "objective" region with comparative, processual and contextual understanding; 2) the personality of a region understood through its specifities, central phenomena; 3) the subjective meaning of a region understood through its identity, sense of place and subjective experience.