GaWC Research Bulletin 242

GaWC logo
  Gateways into GaWC

Note: This Research Bulletin is the English version of RB 241.


Pushing the Taylor Materialist Theses Forward: Concepts for the Nuclear Theory of Metropolis

M. Bańczyk in collaboration with T. Achrem, J. Mroz and D. Śmiejkowski*



The conceptualization of a world city network has reached a point that calls for some finishing but also – for opening of some new horizons. The paper reaffirms methodological primacy of the network paradigm in reasearch on cities and upholds the key findings on cityness, townness and global flows by Peter J. Taylor as articulated most vocally in GaWC RB 177. Pushing forward, it proposes also several clarifications on these findings while some Taylor concepts are rejected and alternative explanations are proposed. This refers mostly to the causal mechanism through which a settlement can reach the global network level. While analysing levels of urban impact, among the so far recognized cityness and townness, the authors introduce additional, third internal perspective of a municipium. The paper concetrates on the very process of metropolization or transformation of a settlement into a network-connected metropolis and proposes a new model of actions based on two theoretical concepts applied in nuclear physics to explain its causes and prerequisites. Subsequently, the paper provides several observations on functions and structure of a metronetwork and distinction between often mixed up phenomena of national micronetworks and regional hierarchies. It proposes also new model for graphic visualization of a world city network combining all three levels of impact.



The outline of the Nuclear Theory of Metropolis (NTM) is in parts our original creation while in other substantial parts it subscribes to the findings of other selected Authors duly cited and acknowledged throughout the text.

The very base and a general starting point for our stipulations and concepts can referred to as “the idea of a global city network” explained and elaborated on in numerous Research Bulletins produced under auspices of and published or re-published within the Globalization and World Cities research project. As our works took several months, we have not been able to consider bulletins published after the RB 229. Out of all these highly valuable papers, we pay special attention to the RB 177 “Cities within Spaces of Flows: Towards a Materialist Understanding of the External Relations of Cities” by Peter J. Taylor. We find it the most pronounced and coherent expression of the strict network approach to the particular theory of a ‘city' and to urban studies as such. Hereinafter we refere to this key source as Taylor Theses or TT. We agree in most points with TT, in some points we intent to reach beyond TT, in other, relatively few points we tend to disagree.

We draw on some findings of other researchers cited in TT, most notably: Castells, Christaller, Friedmann, Jacobs, Sassen, only up to the extent incorporated by TT and never further. Notwithstanding we decided to challenge and reject some of theses proclaimed by these renowned Authors, even if they are strongly upheld by TT.

We venture to say that NTM brings some new theoretical substance, however limited to three points: a) suggesting completions, extensions and further elaborations on Taylor Theses (RB 177), b) providing arguments for rejection of selected theses by Taylor as well as TT-cited Authors, c) providing our own original proposals to replace in part or entirely the rejected theses.

The ultimate goal of the NTM is to help us cognize i.e. get to know and understand the rules that determine, govern or at least influence the development of human settlements such as towns and cities and their mutual relations in the context of globalizing economy of the early XXI century. The most important challenge and the key focus of NTM is to propose a model capable of explaining the process of metropolization defined as a sequence of actions and events causing a so far network-absent settlement (specifically – a regional town) to enter the global space of flows and become a new node in a world city interlocking network.

Our findings result from a complex set of thorough yet sometimes passionate analyses, considerations and discussions held throughout the years 2006-7. These efforts were triggered, continued and coordinated by and within the interdisciplinary think-tank Konsorcjum Metropolia Poznan with an occasional participation of the regional consulting agency IKER, the PATRIMONIUM Foundation, the Poznan University of Economics, the editorial staff and contributors to the strategic city affairs monthly magazine “Metropolia” and several other professionals and passionates all representing the urban environment of the city of Poznan, Poland. In final stage of these activities, we were received by a group of GaWC researchers at Loughborough on an intense and long, mind provoking session led by Mr Peter J. Taylor and participated by Ms Berta Becker (guest), Mr John Harrison, Mr Michael Hoyler, Ms Heike Joens and Ms Kathryn Pain to whom all we are sincerely grateful.

This paper should be considered strictly an initial draft and material for further discussions. Though seeking consecutive improvements and modifications leading to a prospective ultimate accomplishemnt of a coherent theory in rather unknown future, our paper in its present first-draft form does not claim to be anything more than an ambitious try to highlight just another point of view on the phenomenon of global city network.

We propose altogether 70 NTM theses (numbered 1 through 70), organized in seven sections (labeled A through G) that represent the major areas of focus. The additional section (X) provides some indications for a prospective graphic depiction of the metronetwork. With such a rather large number of elements, we suggest to facilitate search though the text by keeping the present absolute numbering, meaning that theses are assigned running individual numbers from 1 to 70 irrespective from the sections, and section symbols just additionally speed up the search.

A. Key Foundations and Terms

  1. Upholding one of the key points of TT (RB 177), we observe that cityness and townness are two genetically different phenomena reflecting two different sets of underlying processes, where cityness exemplifies the network relations. In most settlements they run in parallel which distorts their individual characteristics. The actual type of settlement depends on which of those sets of processes is dominant. The distinction between these fairly opposite types of processes and focusing attention on the so far neglected network relations (cityness) we claim the most important implication for the contemporary urban research. From this point of view, we consider TT (RB 177) and several other GaWC bulletins absolutely fundamental works and an inevitable starting point for further studies. In order to shorten and clarify references, we coin it a term “network paradigm”. Nuclear Theory of Metropolis can be described as further development of Taylor-inspired network paradigm in research on cities.

  2. Coming from TT we acknowledge that the difference between a city and a town is not definable by means of population figures. We challenge the importance of population size in two aspects: a) we deny that causal mechanism of differentiation city-town has anything to do with the population size, b) we claim that the correlation between population and type of settlement is in best case very ambiguous if not utterly insignificant.

  3. When defining network-based cityness, we do radically and consistently reject the territorial dimension of the city itself. In other words, we consider a local hinterland of a city as such equal to zero. Against TT, in order to retain the analytical purity of a model, we see a city as an ideal representation of network processes ONLY. Since network is defined by exchange of processes and NOT territorial gains/losses, this interpretation implies that city as a global node in the interlocking network of procceses is best described by a mathematic concept of intersection, which means it is a point which in turn means it is infinitely small. A network city per se is simultaneously a product of AND a source of global connections between points, NOT: between a point and a plane of field. Each vector representing process flows on a network level runs from one point to another point and is always bidirectional.

  4. In accordance with TT we claim that the dominant scheme of network processes is a win-win scenario. Against TT though, we claim that this is not the only scenario and that there are also processes following the zero-sum-game scheme present in the global city network. In other words, we claim that between the network points (or global nodes, cities), the COMPETITION processes can occur and indeed they do. Network cities fight against one another for certain global assets and competences

  5. TT start with very sharp distinction between a town and a city. This distinction however in some cases further on loses its sharpness and a term ‘city' is used in meaning ‘settlement'. It means such kind of urban establishment that does not have necessarily a character of a network-produced global node in the map of process flows. Even though we intend to maintain the terminology as methodoloigically pure as possible, we find that retrenching the use of the term ‘city' only to the core, network sense will be very difficult given the spread of this word in a common language. We find that similar situations will take place in languages in which we intend to see the network theory translated or co-created. These are for instance: Polish (‘miasto'), German (‘Stadt'), Spanish (‘ciudad”), Portuguese (‘cidade'), French (‘ville'), Italian (‘citta'). From non-Indoeuropean languages that might be important in further studies on city networks, there is a similar situation in Indonesian/Malay (‘ kota '). Therefore, to strictly indicate a settlement with a dominant newtork profile and eliminate confusion as in English (“city”) as in other laguages (as above), we introduce the term ‘metropolis' (sing.) or ‘metropolies' (pl.) We will use this term to refer to settlements that are global nodes in the network of flows of key global resources. We will not say that metropolies will exemplify ONLY the network processes (the concept of cityness form early TT – RB 177). Most of these terms will be alborated on later.

  6. Going further towards a new coherent terminology of processes and phenomena mostly described by TT, we will assume, that: a) a general non-rural concetration of human population with no hint on its city or town profile will be refered to as an ‘urban settlement' or just ‘settlement', b) a global node on a network of flows of key resources and related processes will be refered to as a city or a network city, c) a point having a hierarchical impact on the surrounding plane or region will be referred to as town, d) a built-up area of observable and describable spatial form consisting of a core, a middle zone and a suburbium that are bound together by the combination of social, professional and emotional interactions will be refered to as municipium, e) the total of phenomena, structures, processes that contribute to create a settlement that is connected to other likewise settlements in the world by the flows of key global resources and associated processes while having pronounced municipal level will be referred to as a metropolis, f) the global selection of such settlements forming an interlocking network will be refered to as a metronetwork. Let us stress that a CITY means just the network level, while a METROPOLIS means the entire settlement on all levels provided that it does act as CITY on a network level and fulfills critical conditions as a municipium (on internal level). In other words, having a well developed municipium, it takes only a connection to the global network to become a metropolis. However the major metropolies have regional level also well developed.

  7. Connectivity is the best measure of the global city network known to date. Exchange is its natural and dominant modus operandi. The city network is not organized hierarchically by means of stratification or subordination of one level to another. This does not mean though, that all points of network (cities) function on the same level. Againts TT, we claim that the city network is not perfectly flat and it demonstrates some hierarchical polarizarions, i.e. partial elevations and demotions. We claim there are groups of network cities that are more network-critical than other ones and that are more active and important which means that without those key cities the very functioning of the network might be impaired or even become impossible. We claim there are also groups of cities less important, and their disconnection from the network would not have any major effect on the network (though it would have on the cities in question). The hierarchical polarization of the network is dynamic by nature i.e. the selections of cities constituting network elevations or demotions might drastically change in time.

  8. In order to understand the nuclear theory of metropolis one needs to concentrate on two basic concepts: critical mass and positive feedback. A critical mass means a minimum level of given variable being indispensable for given process to happen. A positive feedback means such interdependance of two processes whereas the result of one stiumulates the source of the other and vice versa. Both phenomena are specifically important in nuclear physics. The critical mass of uranium indispensable for a chain reaction to succeed is about 1kg. Then, the very chain reaction is a typical positive feedback – the result of parted nuclua of uranium causes release of higher number of free neutrons that in turn are able to part more nuclea etc. The important difference bewteen a uranium chain reaction in physics and metropolization is that in physics one atom is brought into play by another atom while a town undergoing metroplization does not make other neighbouring towns follow the way. The positive feedback model in metropolis only explains the mechanism of accuulative character (“snowball effect”) of global position formation while using the mutiple existing nodes of the metronetwork. In other words, a metropolization of Atlanta does not induce a metropolization of New Orleans.

B. Research Methodology

  1. The existence of network is perpetuated by the following crucial processes: flows of all key global resources and specific processes associated with production and/or conversion of ideas / new solutions. However, due to their complexity, intensity, multitude and velocity, we are unable to gauge and follow these crucial processes directly. Therefore, we intent to avoid this limitation by focusing not on the original processes but on the traces they leave in the observable and researchable space. Such a sort of indirect observation bears significant analogy to the approach applied in quantum physics, where we are unable to detect certain particles but we can jugde upon them from the effects of their interactions with other – observable – particles. In our case, i.e. when applied to reaearch on cities, this approach may involve focusing on the formation of global network of offices and divisions of technical oprators running the key global processes. We can assume that the global structure and hierarchy of the establishments of the process operators reflects the structure and hierarchy of the very processes. Going further, advanced producers of services (APS) or key companies supplying the world with advertising, banking, legal and financial services, are certainly fitting well the role of indirect research material for studies on metronetwork. This is exactly the rationale underlying the concept of connectivity in TT. Up to this point we agree with the TT approach exemplified by APS-based connectivity. There is much more uncertainty though with regard to the original pre-selection of cities to be studied. Basically, ANY arbitrary pre-selection puts the whole research at risk, because we might ignore cities of high connectivity that are neither capitals nor seem economically relevant at the first sight and therefore slip out of the pre-selection. From this point of view, it would be better to reject the very concept of pre-selection and DETERMINE the APS-important cities instead through the research on global APS locations. This approach, though methodologically pure, has been proven by Peter Taylor not feasible due to the number of combinations. Bearing that all in mind, we stand on the position of extremely cautious and relatively wide pre-selection that would embrace more than initial number of cities in the first and second GaWC connectivity research.

  2. We claim that APS-based connectivity is only applicable because we have not found so far any better indirect or direct indicator reflecting the flows that constitute the metronetwork. Provided that any direct indicator is found, we shall perceive APS-based connectivity as a second or third choice research option. Nevertheless, we have no certainty that we will ever be able to find direct indicators.

  3. The global APS offices and divisions do not constitute any particularly relevant network that would affect the metronetwork flows significantly stronger or through significantly different way of action that other types of service networks (e.g. mass-media, research institutes, cultural operators, universities, retail chains, etc.). Nor do the APS supply the only products specifically describing the metronetwork. One can easily find global MNC's being serviced by the local APS. What makes APS valuable research material are their location decisions since we assume they locate offices in settlements where there is demand for certain type of services and that APS engagement in these offices is to some extent proportional to the demand. The global network of APS offices themselves does not constitute metronetwork. It only shows where there is demand for types of services associated with global flows that indeed DO constitute the metronetwork.

  4. We find that researching only global APS's (both Taylor connectivity calculations) insufficient and methodologically not recommended. We claim that the APS selection should be based NOT one global (the critical condition of minimum 15 offices in the world) but on individual criteria such as, for instance, turnover basis. If there IS quite high demand for advertising in a city, it is NOT certain that this demand is fully or even adequately met by global APS. There might be a powerful local APS within a global city.

C. City, Town, Municipium - Levels of Urban Impact

  1. We claim that there are THREE levels of socioeconomic activities and interactions generated by human settlements. We will refer to them as urban impact levels. These are: the network level, the regional level and the internal level. The first and the second are well described by TT (cityness and townness). Having said that, we introduce the internal level. We disagree with the suggestion in TT that „cities have some local hinterlands”. We consider “local hinterlands” an environment of processes different to those that create cities AND different to those that create towns. As TT rightly claims, „city is not a large town” since the processes are qualitatively different. The same applies however to the lower stage: the processes running INSIDE a settlement i.e. the exquisite relations and interactions between inhabitants, institutions, public space, authorities etc. cannot be considered network processes. Nor can they be considered regional (hierarchical) processes that are specific to towns. The interior of a settlement is not just a smaller region. In order to refer only to the area of a settlement in the internal sense and the corresponding level of internal processes, we introduce the term ‘municipium'. The internal level of action we will also call ‘municipal'.

  2. The municipal (internal) level of actions affects the very core of a settlement as well as its suburbium. The suburbium is not subject to regional processes but to internal ones. Between the downtown or city center and a subrubium, there is also a space we will call middle zone. In other words, in order to clarify the structure of a municipium (settlement considered only internally) and avoid refering to city or town, we will divide a municipium into three spheres: the core (downtown, city center), the middle zone and the suburbium.

  3. We claim that most municipia feature the same predominant pattern of business locations. The core is home to corporate headquarters, high-value business services, creative services, top hotels and restaurants, high culture institutions, high-end shopping streets, flagship shipping malls and nightlife spots. The middle zone is home to residential areas, regular offices, small and medium enterprises, big stores, technological parks and typical shopping centers. The suburbium is home mostly to factories, warehouses and logistics centers.

  4. Some instances of the real life representation of the municipal or internal level of urban impact (TT – considered “local hinterland of a city”) can be listed as follows: 1) all means of urban transportations especially taxis, streetcars, trains, subways, private cars, motorcycles, bicycles, 2) primary and secondary education 3) part of the telecom operations, esp. fixed lines calls 4) spatial development planning 5) traffic management, 6) security issues, 7) healthcare, 7) everyday cultural events on all levels, 8) everyday leisure and entertainment 9) local elections 10) sport facilities and accessible activities, 11) all cultural facilities, 12) shopping centers 13) road infrastructure 14) airports, bus and train stations, 15) energy supply and management 16) natural environment, 17) quality of local governemnt, 18) most forms of the civic participation and social activities 19) nurturing famous individuals – human icon production, 20) creating spatial uniqueness – architectural icon production 21) all resident nightlife, 22) development of communities and social ties 23) continuity of material inheritance, 24) shaping the resident social capital – habits, temper, special traits, menthality 25) availability of all goods and services, 26) traditions and urban myths, 27) developing and maintaining local identity and pride, 28) the course of all intra-city business processes, 29) teaching activities of universities, 30) limitating or delimitating space for constant events and interactions between the inhabitants, 31) processing of all key global resources i.e.: conversion of candidates into specialists, funds into finances and concepts into new solutions, 32) accumulating the key global resources to critical levels, 33) public space management, 34) location of clusters, 35)creating environment and circumstances of most crucial events in personal life of inhabitants – births, school memories, emotional breaks, weddings and divorces, illnesses, deaths. Municipal or internal impact is generally described by various share of intelectual property, complex and multiple human relations, emotional context, self-governance, spatial context, perseverance as a modus operandi, connection type: point-within-itself

  5. The key factors of the regional level of urban impact are: 1) a territory serviced by the settlement and 2) the relations between the settlement and all points in this territory (region). After TT, a settlement analyzed on this level we will call a (regional) town. The region is organized hierarchically, while the centers of regions or towns feature also a hierarchical order on national scale. The position of a town in this national hierarchy depends on the strength of the impact of town on the region as well as on the size of the serviced territory (region). In contrary to a city, a town has by definition a VERY territorial meaning. We claim that processes on regional level can be best conceptualized by the relation between a point and a plane, not: between two or more points, as it was in case of a city (network level).

  6. We claim that there is a specific pattern of regional flows of key global resources observable on the regional level of urban impact. This is a set of vectors starting from all points on the plane representing a region and pointed towards the town (the dominant settlement of the region). They represent flows of BASIC form of resources: concepts (basic ideas), candidates (basic human resource) and funds (basic form of capital). These vectors are all unidirectional: FROM region TO town. The vectors representing flows of PROCESSED form of resources run in opposite direction: FROM town TO region. These are flows of: specialists (processed form of human resource), finances (processed form of capital) and new solutions (processed form of ideas). We presume that the outward (from town to region) vectors of specialists are rather weak. The outward vectors of new solutions might form concentric rings of affected parts of region which closely resembles the model of innovation diffusion by growth poles (regional towns).

  7. The regional level shows also commodity flows. The vectors run FROM a town OR from other regionally relevant settlements acting either as producers or logistic centers TO the entire plane of the region. They do also run from one town (regional center) to another.

  8. Some instances of the real life representation of the regional level of urban impact (TT – townness) can be listed as follows: 1) most of the national and regional train traffic, 2) bus and car transportation via most roads excluding several transnational highways 3) majority of the telecom operations (phone calls, faxes, sms) 4) production, consumption and transactions of most of the commodities, food, FMCG, business utilities 5) most of the domestic blue collar and low qualified workforce movements 6)locations of factories and logistics centers 7) production, consumption and transactions of simple business services such as wholesales, freight, mechanical repairs, catering 7) cultural events of national importance, 8) specialized scientific, educational, artistic or cultural competitions, contests and likewise prize-based national or regional scale events 9) national political events, pacts, summits, 10) national scale sport events – in Europe: especially regular league soccer games 11) professional congresses, conferences etc. of national importance, 12) activities of movie theaters, themed pubs and bars, concert halls and likewise pop-cultural operators 13) locations and activities of the national level public institutions, governmental and para-governmental organizations or their special divisions 14) locations of the knowledge-non-intensive specialized BPO (business process operation) centers such as accounting centers, 15) locations of sales offices of major corporate players, 16) size of the territory serviced by the key universities (measured by origins of students), 17) regional pride and identity including dialects, customs etc., 18) location of the HQs of domestic corporations, 19) simple money transfers, deposit banking operations, cash loans and regular saving account capitalization, 20) migrations and population trends, 21) domestic tourism traffic patterns, 22) domestic direct investment inflows, 23) historical profile and importance, 24) most international and domestic operations on industrial and commercial space rental and related real estate transactions, 25) diffusion of innovation to the province – sending trained specialists or exporting some minor solutions back to the region, 26) attracting basic for of key global resources from the region i.e. absorbing unqualified candidates to companies, absorbing money from the region, absorbing concepts created in a region but not well developed and infeasible, 27) location and activities of the national and regional administration including all government establishments and agencies. Regional impact is in general described by low to medium share of intellectual property, zero-sum scenario, competition against other regional towns as the main modus operandi and mostly unidirectional connection type: region-to-point.

  9. Some instances of the real life representation of the network level of urban impact (TT – cityness) can be listed as follows: 1) regular passenger air traffic (airlines are by far the most important means of transportation on network level), 2) scientific activities of top universities and research institutes demonstrating intense and sustained international co-operation, 3) high volume transnational data transmission, 4) transactions of designs, brands, licences, trademarks, copyrights and other intellectual properties corresponding to the creative industries (as defined by DCMS, UK), 5) corporate know-how transfers (within multinational companies), 6)locations of HQs by multinational companies, 7) production, consumption and transactions of knowledge intensive business services such as technological innovations, 7) cultural mega-events of transnational importance, 8) scientific citations (SCI) especially with regard to co-authorship, 9) big international political events, treaties, summits, 10) world-scale sport events and exhibitions, such as olympics or regular expos (not all expos), 11) professional congresses, conferences etc. of transnational importance, 12) activities of the high-end, internationally recognized institutions of the core art fields (visual arts, performing arts, heritage) such as theaters, operas, museums, 13) locations and activities of the international governmental and para-governmental organizations such as UN, OECD etc., 14) locations of HQs of the globally relevant NGOs, 15) locations of the main national media, especially – TV stations and biggest national daily papers, 16) locations and activities of companies operating in the creative industries (as defined by DCMS UK), especially – design, architecture, software, 17) sophisticated and complex long distance financial operations such as futures and options on overseas stock exchanges, non-cash mergers and aquisitions, hegding, private banking, private equity management services etc., 19) production and consumption of individualized insurance, pensions, mortgage, credit instruments etc. and capital-related products for personal use, 20) foreign tourism traffic patterns, 21) locations and activities of the domestic companies exporting their services nationwide or even worldwide, 22) foreign direct investment inflows and outflows plus domestic direct investment outflows, 23) all significant operations in the international office space rental, 24) most movements of domestic and international talents, of highly qualified specialists, white collars, CEO's and other top management professionals often accompanied by high absorption of domestic and international university graduates, sometimes also - low-qualified internationals (immigrants, exiles), 25) generally visible presence and diversity of foreign citizens 26) locations and to some extent activities of the advanced producers of services in the TT sense (APSs). Network impact is in general described by high share of intelectual property, mostly a win-win scenario, exchange and cooperation as the main modus operandi and a bidirectional connection type: point-to-point.

  10. It is interesting to point out that while there is a clear qualitative difference between all three urban impact levels, the network level and municipal level are more linked one to another than either of them to the regional level. The municipium is the very core of a metropolis while the network is nothing but an intense and multidirectional exchange between the municipia in the world. Municipium gives a metropolis its physical presence and social capital profile. If a municipium is highly developed, getting connected to the global metronetwork is facilitated. And the other way round – global connectivity only improves the municipium – its spatial form, atmosphere, social profile. The key global resources traveling in the global network exchange eventually land in one or another municipium. These two perspectives – internal and global – are closely interlinked. The regional (hierarchical) urban impact (townness) does not match this municipium-network paradigm. It is a distinct gravity-like mechanism that features some characteristics of the guardian type of social reproduction, i.e. it is not so far from the concept of state with all its territoriality, hierarchy and dominance. The regional impact cannot be neglected since it may highly reinforce the municipium's ability to achieve critical mass of resources. However, since we can find global metropolies that demonstrate almost zero regional impact, we have to conclude that regional power is not indispensable to achieve global connectivity. Moreover, the regional level of urban impact does not appear to be just a non-essential middle part of a metropolis activity structure. It appears to be a remnant of an alternative and to some extent – competing form of developmental mechanism of human settlements. After all, for ANY kind of flows there must be some basic provision: energy, people, capital, space, infrastructure. This might imply that a metropolis CANNOT simultaneously increase its regional power and network connections. Then, if there is a choice, it can be easily argumented that regional power should be sacrificed to global connectivity because it is obvious that global metropolies do significantly better than regional centers. In the past few hundreds of years when the world continued to fall into separated and only more retrenched pieces called states and connectivity was a dead word, regional hierarchical path of town development might have been the only reasonable course of action. It is not anymore, though.

D. MetropOLIs

  1. There are two conditions for a settlement to be classified as a metropolis: 1) the clearly dominant global network profile (TT - cityness), 2) the highly developed municipium, whereas the level of development will be described with regard to both spatial and processual dimensions. We are also aware that a vast majority of metroplies will feature highly pronounced level of regional power but we do not see regional power as a necessary metropolis property. There are scarce though significant case of global cities with regional impact close to none. In practice though, when using a term ‘metropolis' we will mean settlements displaying all three levels: network (city, global node), regional (hierarchical, town) and internal (municipium). The most important metropolis-wise will be the network interactions while still critical will be municipal level.

  2. A typical activity of a metropolis combines all three levels of urban impact: global network, regional and internal. A metropolis attracts and absorbs resources from the region, converts and processes them, then a relatively small portion of resources is expedited in processed form (for instance candidates converted to specialists – human resource) back to the region, while the large portion of resources is kept inside the municipium which creates an appropriate critical mass of resources necessary to produce more complex, innovative and sophisticated new solutions. New solutions are mainly expedited to other metroplies i.e. one metropolis acting as producer of new solutions sends them TO the metronetwork. At the same time the metropolis in question receives foreign new solutions FROM the metronetwork (i.e. from other metropolies producing new solutions). New solution, as we said, is a complex processed form of the key global resource of ideas. By definition, the ideas (unlike commodities) even if supplied to a new receiver remain still with the original producer (sender). This means that the exchange mechanism of new solutions that runs constantly in the metronetwork leads to significant condensation of new solutions in the participating metropolies because they collect the self-produced new solutions and receive the new ones from the metronetwork. This is how a high concentration of new solutions in a metropolis is created. It is utilized mainly in municipium (on metropolis own, internal level and for the sake of it) but it also radiates to the region. The diffusion of innovative new solutions from a important town or city to the region has been well known for some time under the concept of growth poles. This concept however has so far lacked the network perspective. Our contribution is to claim that a growth pole is at least partially (if not – very significantly) powered from outside, i.e. from the metronetwork flows. This is a new interpretation of growth pole through the network paradigm introduced by TT and upheld by us. The diffusion of new solutions from a metropolis to the region is more-less homogenic in all outward directions, however the flow of solutions to the region is considerably weaker than the metronetwork flows.

  3. A metropolis can also be described through the parameter of density. The term is commonly used and refers mostly to the spatial density of the built-up area (municipium) fo a settlement. However, we intend to introduce the other, if not the most relevant, aspect of density of a metropolis. It can be called a processual density. We will define it as the maximal number of processes and/or interactions involving metropolis actors (people, indstitutions) that take place or could possibly take place at the same point of time within a given municipium.

  4. We claim that being a metropolis is in general better than being a non-metropolis. We claim that metropolis makes in general a better living habitat than other forms of human settlement, especially – better than regional town. It is not only due to its economic supremacy manifesting through higher personal income, purchasing power or growth potential (illimitable because of a metronetwork). It is also (if not more significanlty) because metropolies offer more possibilities i.e. wider spectrum of options in practically all fields of choice in human life. This multilayered spectrum of choice is a direct consequence of high processual density that is typical for a metropolis.

  5. Since a given lifetime of an metropolis inhabitant is filled with higher number of variables (events, options, etc.) than a given lifetime of a town-dweller, the relative perception of time lived ex post appears longer than otherwise. In other words, even though living in a metropolis seems hectic and passing by rapidly, indeed it prolongs every unit of time because the same hours that would be empty elsewhere in a metropolis are inscribed in memory with large number of events which makes them appear richer and longer.

  6. We claim that a metropolis can make a starting point on the map of commodity flows. However these flows are NOT important on the network level. The site of action of commodity flows is almost always a suburbium. A metropolis as a starting point of commodity flows uses either its internal or – more often – its regional level (hierarchical).

  7. The simple services are by definition local. Nobody travels to other distant towns to have a drink poured or hair combed unless he or she is another ipersonification of Elvis Presley. Simple services, even though scarcely creative and little value-enhancing not to mention their limitation in productivity growth, are also very important to a metropolis. These services allow for recurrent circulation of the capital resources pumped into a metropolis (municipium level) from metronetwork. Otherwise, even at high intesity of the exchange of a metropolis with the metronetwork, the situation could approach the picture of ‘having a pipeline but lacking a tap'.

  8. A metropolis can be also described as a producer of new solutions of global quality. It absorbs the resources, mainly in basic form, from the region, processes the resources in the municipiun and then sends them back in proceesed form – mainly as new solutions – to the metronetwork while diffusing some to the region, too. The form of products – goods or services – that act as driver of flows has not much importance since within the commodities there is a certain ammount of frozen processed idea (solution) present. A metropolis also absorbs resources from metronetwork.

E. Metronetwork

  1. We introduce the term ‘metronetwork'. The meaning of it is very close to ‘global network' or ‘interlocking network' frequently used in GaWC bulletins and of course in TT. We do however make two clear provisions. First, we assume the influence of national boundaries and political systems is utterly negligible in this regard and therefore w exclude state- or guardian-related issues from metronetwork concept. Second, by using the specific ‘metro' prefix, we point at the more complex concept of a ‘metropolis' (defined and elaborated on in further parts), to stress the fact that all settlements functioning as points on city network function also otherwise on two different levels of socioeconomic activity (regional and municipal) which makes them METROpolies and that these levels mutually interefere.

  2. In TT there is no clear division between the critical types of flows that constitute the network and those that play lesser role and just happen to exist in the city space of flows. Many flow types were mentioned or hinted upon during discussions following TT. This includes commodity flows, invoicing and payments as well as IT or marketing solutions being disseminated worldwide through the global structures of advanced producers of services (ASP). We claim clearly that there are three types of flows that constitute the metronetwork. These are flows of the key global resources: capital, people and ideas. The flows of ideas are by far the most important.

  3. Each of the three types of key global resources can assume two forms: Basic and processed. We will name the human resorce in basic form: candidates, in processed form: specialists. The basic capital resource will be named funds, in the processed form - finances. The last though most important global resource of ideas in basic form will appear as concepts while in processed form – as new solutions.

  4. During the process of conversion from the basic form to the processed one, all three global resources remain in the mutual interrelations. Conversion of one basic resource into processed one enables and facilitates the processing of other resources. The nature of resource conversion bears resemblance to the phenomenon of a chain reaction.

  5. We claim that flows of commodities have little significance to the metronetwork in which we differ from TT. These flows, expecially those of certain goods, are designed and managed by metronetwork, but they neither contribute to network development nor follow exactly the network connection patterns. The global economy of early XXI century features deepening separation of tangible and intangible processes in product development. The intangible part of product value creation comprising such processes as research, product conceptualization and design, brand development and management are less and less often connected to the tangible part, i.e. manufacturing, assorting or distribution. This separation of processes means different locations, operators and owners. The location of intangible processes is nowadays not even significantly influenced by the location of production sites or trajectories of commodity flows. What is critical to the metronetwork is the intangible part of processes, not the whole commodity flows. In other words, the metronetwork determines the commodity flows while it is not determined by them. Nevertheless, the metronetwork connections are used in perpetuating commodity flows i.e. certain proportion of all commodity flows goes along the metronetwork.

  6. We claim that commodity flows in metronetwork are not only insignificant, they can turn into a highly elusive factor. For instance, suppose we analyze a case of an inhabitant of, say, Gdansk, Poland, travelling to Warsaw and purchasing the newest model of Nike sneakers produced in Bangladesh. It appears as a following flow: human and money transfer from Gdansk to Warsaw, product flow from Bangladesh to Warsaw. In fact it means a flow of idea (new solution) from New York to Gdansk with a little help of Warsaw. The flows of commodities are misleading if not mischeavious as within these flows there are much more important flows frozen and hidden. Nuclear theory of metropolis (NTM) requires a surgical dissection of commodity flows and a precise extraction of crucial flows of ideas from the superficial picture of commodities. It calls upon the concept of knowledge-intensive businesses or – even more clearly – upon the creative sectors.

  7. The materiality or tangibility of product resulting in a comonly known distinction bewteen goods and services is quite insignificant and inadequate for the nuclear theory of metropolis. What is way more important and applicable is the analysis of processes in the value creation, i.e. measuring the contribution of intellectual porperties or creative processes to the product creation on every stage. While some professional services are almost pure form of creative processes, the other ones – simple services – are powered by creativity drastically less than some of the mass manufactured goods. During the production processes, the creative (intellectual) value, is actually frozen and multipied by means of mass distribution. Similar phenomenon can be detected in the production of services but the multiplication ratio is much smaller.

  8. Metropolies differ by the way they interact with the metronetwork, i.e. whether they mainly power the network or are powered from it. In order to clearly describe this phenomenon we introduce the concept of a network account of a metropolis (NAM). We will define it as a difference between the total ammount of key global resources supplied to the metronetwork and the total amount of key global resources received form the metronetwork.

  9. Using the criterion of NAM, the entire metronetwork can be divided into three groups of metropolies: net-dependant, net-supporting and net-neutral. These groups correspond respectively to: negative NAM, positive NAM and NAM equal to zero (balanced). This is of course only analytical concept while the very situation of a metropolis is very dynamic and the NAM can present frequent and sharp changes.

  10. The total metronetwork account (TMA) must be equal in value and opposite in sign to the sum of all individual network accounts of metropolies (NAMs). In other words, a net leakage of key global resources from the metronetwork or a net injection of key global resources into the metronetwork must be exactly explainable by a net leakage or net injection of resources to the region, mucicipium or micronetwork (later described) attributable to the corresponding net-dependant or net-supporting metropolies.

F. Metropolization

  1. The process in which an existing settlement (acting as regional town or only as a municipium) reaches the global network level we will call a metropolization. In other words, we presume that most metropolis-candidate settlements have already developed some municipal level, and some of them also regional level of activities so a critical factor potent of transforming a settlement into a metropolis can be defined as Metronewtrok connectivity. There is some analogy between so defined metropolization and a process of transformation from a regional town to a city in Taylor sense (TT). However, already on the very mechamism of transofrmation, not to mention its triggering factors, we differ significantly. Since cityness and townness are rightly defined in TT as types of processes founded on different activities, conversion from a town to a city entails conversion of underlying processes. In other words, if a ‘town' is a territorial zero-sum-game hierarchy phenomenon, and a ‘city' is a win-win, non-hierarchical network phenomenon, then a transformation means there is no more townness in the resulting settlement that starts acting as a city. Such a mechanism would sooner or later lead to a very polarized urban reality featuring only pure forms of towns or cities in TT sense. Which does not appear to happen. According to the core TT (RB 177), most settlements harbor both types of process – townness and cities – at the same time. This is however contradictory to the very consequence of TT which predicts that conversion of town to city does NOT retain the town characteristics. According to TT, a city is not a surplus of cityness over the remaining townness but it is indeed a qualitative change in the settlement as such. In order to avoid this contradiction we propose a different conversion path. We see the critical importance in metropolization defined NOT as evolution from a town (regional center) to a city (point of a network), but from a town (or even municipium) to a metropolis, i.e. to a complete organism acting on global network level but ALSO retaining the regional and/or municipal level. Such an understanding of metropolization or key urban transformation allows for focusing attention on network perspective as proposed by TT while it also provides an explanation of the empirically observable co-existence of network and regional processes in the contemporary settlements.

  2. We reject the thesis by Jane Jacobs upheld by TT that a “new work” is a decisive driver of metamorphosis from a town to a city. Generation of a new work meaning new business activity presumably in a so far unutilized sector or industry of a town is certainly positive for the economy. Nevertheless, by no means does it make a town approach the network perspective. Moreover, beaering in mind that TT describe a transformation from town to city and NOT a formation of incremental cityness over existing townness, generation of a new work should cause a conversion of a territorial organism based on hierarchy and competition into a network organism based on exchange and cooperation.We see no reason why it should happen. We claim that a town that experienced even multiple growth of ‘new work' in Jacobsean-Taylor sense (including new products, services, technologies, etc.), caeteris paribus still remains a TOWN, yet somewhat more diversified.

  3. During the IKER – GaWC Session (Loughborough, 13.06.2007) Peter Taylor suggested that „town” is indeed an „uncomplicated city”. Early TT clearly state that between town and city there is no simple quantitative or scale difference! We remain faithful to this early stipulation. We claim that a network-based exchange versus a regional hierarchy and competition is not a matter of complication or complexity. These are simply different syndroms.

  4. After a discussion on long disputed pros and cons of diversifications and specializations in world economies we still fail to see relevant arguments in favor of diversification of an urban economy a priori. There are numerous case studies from the world economies where even a very carefully planned diversifications proved to be counterproductive. Why should diversification of the urban portfolio of industries be assumed invariably positive? It is probable that a service portfolio available in a town can grow almost infinitely to the good of the inhabitants. However, it can be argued that having access to a very large selection of most potential service sectors, some of them one may never need, could be less beneficial to the inhabitants than having access to smaller selection of service sectors but with great degree of competition meaning more providers in each service sector. Going further, the same diversification-based approach cannot be applied to the sectors of manufactured goods. Why should a town be able to produce everything? Diversification brings some safety but greatly dimminishes chances of building and unitilizing a critical mass, it debilitates the growth pole factor, reduces possibility of reaching excellent level in any sector and highly complicates the perspective of building specialization clusters. It is due to the inevitable dilution of resources. Of course, specialization in one sector does not require having only one sector. Good instances are film industry of Los Angeles, San Francisco Bay with its Silicon Valley or financial-headquarters cluster of Frankfurt am Main which proves that urban economy can be to some extent multidisciplinary and still somehow specialized. Altogether it does not explain at all how a regional town should turn into a network-connected global city just because its economy is getting more diversified. Such a town will only become a diversified town. No more.

  5. We reject another thesis by Jane Jacobs upheld by TT, that a new work, be it manufactured foods, services or technologies, result from import substitution. We watch the growth of several Polish cities approaching or already having relatively strong regional position but with very scarce or merely none trace of a world city formation processes. We are unable to name ANY single ‘new work' that is now supplied from imports and that if generated domestically - would make these cities more network-connected. The flows of commodities, partially described earlier show rather an opposite rule. For instance, if some tropical fruits, say rambutans, are suddenly successfully planted in Wroclaw, Poland then it would only DECREASE and not INCREASE the commodity flow between Wroclaw and Jakarta, Indonesia, rambutans' homeland. This example to some extent works also in the world of professional services. What stimulates more exchange with the global cities or metronetwork? Having agencies like TBWA or Norman Fosters Architects and Partners OR substituting them by even most skilled local counterparts? It speaks very high of town's potential to have local professionals on such level but it does not increase the town's connectivity at all. We presume that the import substitution has rather NEGATIVE impact on network connectivity. Moreover, there is no proof that new work, even if not increasing connectivity, comes only from import substitution. This reasoning is based on erroneous assumption that there is a given and wide spectrum of existing needs or – a given diversified structure of demand for goods and services in town - and the only difference results from shifts in supply, whether it is domestic or comes from imports. This is clearly wrong. The structure and scale of demand for services and goods in town undergoes constant changes and demand for some types of high services develops only in SOME settlements, while in other ones it never starts to exist. Then, if there is no demand for some classes of goods and services, there is no imports of them. The non-existing imports cannot be substituted. The key problem and one of the key findings of our theses is that the regional towns (and in general - settlements) disconnected from the metronetwork very often lack demand for certain professional services especially from B2B or personal amusement (advanced leisure) zone as well as they show no demand for certain high-end goods. This in turn proves to be a very important and negative location factor for headquarters of companies and a discouraging factor for a new qualified labor (specialists) that in our times of personal freedom and hedonism seems specifically succeptible to the threat of being bored by too narrow offer of city wonders. These two negative factors combine to keep town economy away from the potential nucleus of new demand that might in turn breed new supply. There is no import to be substituted and no one who could substitute it even if there was a need. The town remains disconnected or even turns autarkic. This phenomenon can be explained by the well known mechanism of feedback. This is of course a feedback of positive type (the more – the more) which means it drives phenomenon towards the extreme, not towards the equilibrium. No demand for high-end products and services – no companies eager to supply them. No potential suppliers – no new actors willing to move in. No headquarters. No process flows. No APSs. No connectivity. The Jacobsean concept of import substitution proves wrong because it expects a town to seek to substitute imports of a city, a city this town IS NOT at the moment. In other words, it assumes substituting imports that a town does not HAVE because there is no demand for them. The feedback effect in this case brings an accumulating autarkia. We claim though that the same mechanism of positive feedback is critical for metropolization.

  6. In order to make a town become a metropolis – meaning: a town reaching network connection – we need to use the mechanism of positive feedback but divert it to the opposite direction. We need to divert the feedback from autarkia towards multiple external interaction. This means stimulation of a chain reaction: at the beginning there must be the first explosion – either a demand for global quality without an existing supply or a supply of global quality without an existing demand in town. Such an appearance of new demand or supply of high-end, global class goods and/or services can be interpreted as initial investment on the rewarding conversion path from regional town to a metropolis. The demand or supply surpassing certain critical mass and persisitng voer critical period of time must inevitably provoke a gradually intesifying response meaning new suppliers, new labor coming in, new headquarters. Then, a typical snowball effect follows, APS establishments multiply and we see another world city or a metropolis growing. We must observe that a contemporary rise of a global city is almost invariably a fast, accelerating and ardous process rather than long mild evolution. Such an observation confirms the chain-reaction (or positive feedback) approach.

  7. Unlike TT (RB 175) we claim that a complete service self-sufficiency of a town is NOT necessarily beneficial or even IS mostly undesired because it increases the risk of autarkia-aimed feedback mechanism in a town and reduces stimuli to seek contact with the outside world.

  8. We claim that metropolization being a process leading town population to start living in a metropolis is beneficial to the most inhabitants i.e. to those normally motivated. By normally motivated we understand individuals who prefer a longer and more interesting life to shorter and less interesting.

  9. We see two conditions critical for a successful metropolization: 1) there must be a critical mass of resources available in a municipium, 2) a prime stimulus must emerge, being a first sparkle of the metropolization chain reaction. This stimulus works as a triggering factor, like a hygroscopic particle that launches a formation of the first drops of rain. This factor we will call a nucleus.

  10. A critical mass of resources in a municipium is necessary for the first phase of metropolization which is before the flows from the metronetwork and from the region are attracted and absorbed.

  11. We have said that nucleus is the main trigger or stimulant of metropolization. It is an economically relevant phenomenon associated with a settlement (town), that EITHER creates global resources in processed form, especially a new solution of global class that could be expedited to the metronetwork and absorbed globally OR requires global resources (esp. new solutions) available only in the metronetwork OR simulteanously creates globally absorbable new solution and requires globally available resources. In other words, nucleus is the event, process or object that forces a town to connect to the metronetwork and to remain connected.

  12. The self-suficiency of a settlement is neither a sign of metropolis nor of a town. The key and adequate concept to undestand the self-sufficiency lays beyond the type of settlement and can be described as a balance between the needs and supplied products, especially service products we can name solutions. The error of former approaches to this topic stems from the fact that only the supply side was analyzed while the spectrum of needs (demand side) was treated as granted and constant. It was consequently excluded from the consideration. Therefore the self-sufficiency of a town was erroneously understood as the ability of a town to meet the assumed constans of needs (presumably high and sophisticated). In fact, the self-sufficiency means ONLY that the level of needs in a town is at the moment equal to the level of offered solutions or – that the level of solutions is higher. It does not mean that the level itself is high or low. We therefore reject one of the key claims of TT (e.g. RB 175) that a service self-sufficiency is a pre-condition of a (gateway) city. If the level at which demand-supply balance was achieved is LOW, then a town remains a town no matter how perfectly self-sufficient it is. The self-sufficiency of a town does not even prove its strength on the regional (hierarchical) level. It is indeed an inadequate concept.

  13. In order to improve the descriptive precision of the criterion formerly known as service self-sufficiency of a town, we introduce the concept of vertical account of needs and offers (VNA). By vertical we mean that the VNA concept refers to the number of diverse TYPES of needs and offers which in turn relates to the level of sophistication of both demand and supply with regard to professional services. We do not mean the intensity of need i.e. the number of people of enterprises presenting a given need or offer.

  14. A balanced VNA will mean that a town offers such a spectrum of goods and services that meets the current needs of a municipium and a region. The unbalanced VNA will mean that it does not. The unbalanced VNA exists in two forms. Positive VNA means a town offers goods or services for that there is so far no demand present in municipium nor in a region. Negative VNA means there are needs in the municipium or a region that cannot be met by the offer of a town.

  15. The process of metropolization can be triggered only if the VNA is forced out of its equilibrium position. It takes place through nucleus. Otherwise, supposed constantly balanced VNA, a metropolization will not take place.

  16. A nucleus emerges and is powered in the initial phase by the resources of municipium.

  17. A nucleus causes increased utilization of certain resources and increased demand for them. An increase in town activity takes place, initially – still within the municipium.

  18. Subsequently, after the critical level of activity is surpassed, a chain reaction commences. The chain reaction runs through the metronetwork connection however it can also involve regional hierarchy structures and national micronetwork of towns.

  19. A typical example of metropolization nucleus can be seen found in a formation of cluster of world class products, nowadays predominantly – service products. A cluster does not need to be industry-specific. It may be based on horizontal (i.e. industry-type) criteria: Sillicon Valley of San Francisco Bay, financial services of Frankfurt, selected sorts of tourism and leisure in Rio de Janeiro, jet manufacturing in Seattle, transportation hub in Hamburg, etc. Yet a cluster can also be determined by using the vertical analysis of corporate functions. The most widely present type of clusters based on vertical criteria is actually a headquarters cluster. While theoretically equivalent headquarters of companies are present in many metropolies scattered around the globe (for instance – country management offices of Citibank), they tend to play different role in different places. In some cases they are an accompaning phenomenon of the otherwise strong service cluster specialization (Milan, Barcelona) while in other cases they actually CONSTITUTE an otherwise non-existent cluster of top corporate offices (Singapore, Warsaw) transforming their hosting metropolies into the very command and control centers in Friedmann's sense.

  20. TT (RB 177) stress that a city is NOT product of an evolution of a town, moreover - that the rise of town has been chronologically preceded by the existence of cities. This notion is of high importance bearing in mind the key challenge of our theory which is to research, understand, model and eventually control the process of managing towns toward a newtork city perspective. Coming from the TT point of view, we must conclude that analyzing history of cities can make no relevant contribution to this research since if cityness predates townness then historical transformations from towns to cities could not have been observed in the past. We further claim that the only hints and important conclusions can be drawn from studies on selected cities where cityness has significantly changed either way. We claim that the research value of such cases is subject to their age. The more contemporary, the better. In general, with regard to metropolization processes, history will teach us nothing.

  21. Metropolization, especially a strategic metropolization defined as a delibarate use of a carefully planned city strategy in order to transform a regional town into an open network-connected metropolis is a new phenomenon, a challenge pressed by the globalization era. It has no counterpart nor close reference in the economic history. Bearing in mind that, as TT rightly claim „cities predate towns”, metropolization appears even a process somehwat reverse, a return to a more basic mechanism in the world of city development. Yet indeed, it is an entirely new phenomenon. No other period in history could possibly rival the present day with regard to the number, size and economic power of the cities that are functioning at the same time in one globe while the technology allows for a practical co-habitance of all world cities in one economic environment. At the same time the average purchasing power is at its peak of all times. These all conditions mean that at no other point of time an existing or potential global network of metropolies (metronetwork) could have been even distantly comparable in width, reach, complexity and the population size involved with the exisitng or potential metronetwork of the beginning of XXI century. As the underlying processes accelerate (reduction of barriers for global flows of resources, constant human migration to cities, rise in GDP, population growth), we can presume that the metronetwork will become increasingly complex, wide and important. We might be watching the beginnings of a new era – an era of global network-powered metropolies.

G. Regional Hierarchies and Micronetworks

  1. TT takes up the issue of state-related influence on the city network which involves the interference between two major ways of social reproduction: guardian/taker and producer/trader. However, the selection of important settlements within one state, called in TT a national city system, is interpreted ambivalently, once as a mini-network, at the other times – as a hierarchy of regional towns. We claim redically that a national city network is an utterly DISTINCT phenomenon from a national hierarchy of towns EVEN if technically both groups contain exactly the same selection of settlements.

  2. The hierarchy of towns is a pattern of ordering the settlements located on a given territory by their strength of regional impacts. This hierarchy is an examplification - and at the same time a natural consequence - of the regional processes (townness). It is mostly limited to a territory of one country or state and therefore it is often called a national hierarchy of towns. There ARE however towns of transnational regional impact.

  3. The domestic network-organized groups of cities and towns, whereas a vast majority demonstrates significant connectivity limited only to a territory of one state or country, we will call (national) micronetworks.

  4. Many micronetworks operate such way that one their elements (towns) is part of the metronetwork at the same time. Since all other nodes of micronetwork are only connected with one another and not with the outside world, all the metronetwork-bound flows originating in micronetwork must come through this double node (a city connected to both national and global network). Such form of national micronetwork having just one connection to the global metronetwork we will call a loop.

  5. A metropolis acting as the common double node of both micronetwork and metronetwork we will call a gateway.

  6. We claim that the relationship between the national hierarchy of towns (regional level, townness) and the national micronetwork is ambigous. The national town number one, or a settlement featuring biggest regional power in a country, does not have to be gateway and gateway does not have to be national town number one. Town is a matter of territorial hierarchy and gateway is a matter of connectivity.

  7. Since the metronetwork overlaps with the micronetwork and metropolies operate also on municipal (internal) and regional level, the total account of flows between all points of the metronetwork (all metropolies) does not necessarily have to be equal to zero.

  8. The contemporary metronetwork powered by globalization era is merely too young phenomenon to allow for describing the rules determining the relationship between micronetwork and metronetwork. It seems that a regional town, once metronetwork-connected (metropolized) ascends in the national town hierarchy, yet there is no proof it must reach the absolute peak position in national hierarchy of regional power. The opposite case though (the influence of increasing regional power on network connectivity) must be seen in at least two separate options. If the settlement already IS network-connected, i.e. if it is an existing global metropolis, the otherwise induced increase in regional power results in intensified flows along existing global connections without inducing new connections. The other case takes place if a town is not network-connected at first. Then the rise in regional power probably somewhat facilitates meeting pre-conditions for metropolization since it helps achieve the critical mass of resources through the absorbtion from larger territory. However the increased regional power does not appear to form any sort of nucleus being a critical condition of metropolization and its triggering factor. It seems that in general, there is more impact of global connectivity upon hierarchy of regional power, than the other way round. It has not been proven yet that the latter impact exists at all.

  9. We claim that also for a state (country) having metropolies is better than not having them. Moreover, we claim that having more metropolies is in general better than having less metropolies. We imply that metropolies pose conditions for a better, dramatically more innovative, efficient and enriching organization of development for the entire society, not only for those living in metropolies. In other words, we claim that the short- and long-run benefits of metropolies are felt also by those living in a non-metropolies as well as by the state perceived as a souvereign entity.

X. Visualization

The concept of triple urban impact can be graphically presented through the cone model. A settlement is just a point on the map. A metropolis we will depict as a cone standing ABOVE the map and resting on it.

The size of its base represents the regional power. The more powerful a town, the larger territory lays under its influence. More precisely, the graphic representation of the regional processes as described earlier would need a very large number of unidirectional vectors running from all points within the cone base toward the center of the cone base.

Now, the exellence of the municipal or internal level would be represented by the height of the cone or the vertical distance from the base center to the peak. The most important level – global networkness – would be depicted as a radial pattern of lines connecting the peak of the cone with other peaks of cones representing other global metropolies resting on other parts of the map.

The metronetwork connections do not run all on exactly the same level. Some peaks are higher, the other ones are lower. This difference in heights corresponds with the fragments of network polarizations (elevations and demotions) described earlier as global metropolies are not ideally equal and the whole metronetwork is not totally devoid of hierchical inluences. The higher peak of cone, the higher connection line. The higher network connection line, the more intense and better quality the global exchange of key resources. In other words, the intensity of network exchange depends on the height of the cone. As said before, the height represents the level of municipium quality. Which seems to confirm the earlier assumptions: the better developed municipium (high cone), the better a metropolis prepared for fast and intense global flows (high connection lines). Once again, the municipal and network levels of urban impact combine harmonically.

The last parameter to be explained is the steepness of the cone slope, i.e. the steepness of a triangle that results from a vertical intersection of the cone. It is simply an expression of relation between the strength of combined levels municipium-network AND the regional strength. The flatter cone, the more regional impact, the lower cone, the worse developed municipium and less intense network connections. In other words, long and slim cones wll represent metropolies of crucial global role yet little regional townness activity (Singapore, Geneva), while short and flat cones will represent those metropolies of significant position in regional hierarchy but little developed municipium and scarce network connections (Yogyakarta, Minsk). Metropolies characterized by equally strong municipium-network level AND regional impact (London, Chicago), will be depicted as high cones of 45 deg. angle at the base in vertical intersection.



Cushman & Wakefield Healey & Baker: European Cities Monitor 2004, 2005, 2006

UBS AG : Prices and Earnings 2003

Mercer Human Resource Consulting: Cost of Living Survey 2005

Mercer Human Resource Consulting: Quality of Living Survey 2005

Anholt / Global Market Insite: City Brands Index 2005, 2006

UK Dept. for Culture, Media and Sports: Creative Indistries Mapping Document 1998, 2001

European Commission: State of European Cities Report 2007

European Commission DG Regional Policy / Eurostat:


1. GaWC Research Bulletin 229 (A) : 'Cities and States: An Elemental Reinterpretation of their Origins and Relations'
(P.J. Taylor)

2. GaWC Research Bulletin 225 (A) : 'The Urban Network Transformation: Planning City-Regions in the New Globalisation Wave'
(K. Pain)

3. GaWC Research Bulletin 224 (A) : 'Balancing London? A Preliminary Investigation of the 'Core Cities' and 'Northern Way' Spatial Policy Initiatives Using Multi-City Corporate and Commercial Law Firms'
(P.J. Taylor, M. Hoyler, D.M. Evans and J. Harrison)

4. GaWC Research Bulletin 221 (A) : 'A New Theoretical Basis for Global-City Research: From Structures and Networks to Multiplicities and Events'
(R.G. Smith and M.A. Doel)

5. GaWC Research Bulletin 219 (Z) : 'From Competitive Regions to Competitive City-Regions: A New Orthodoxy, But Some Old Mistakes'
(J. Harrison)

6. GaWC Research Bulletin 217 (Z) : 'Analysing the Changing Landscape of European Financial Centres: The Role of Financial Products and the Case of Amsterdam'
(J.R. Faulconbridge, E. Engelen, M. Hoyler and J.V. Beaverstock)

7. GaWC Research Bulletin 216 (A) : 'Resurgent European Cities?'
(I. Turok and V. Mykhnenko)

8. GaWC Research Bulletin 213 (Z) : 'Constructing New Regional Geographies: Territories, Networks and Relational Regions'
(J. Harrison)

9. GaWC Research Bulletin 210 (Z) : 'Global Research Centres: Dynamics 1996-2004, Networks 2002-2004. An Analysis Based on Bibliometric Indicators'
(C.W. Matthiessen, A.W. Schwarz and S. Find)

10. GaWC Research Bulletin 209 (A) : 'From North-South to Global South: A Geohistorical Investigation Using Airline Routes and Travel, 1970-2005'
(P.J. Taylor, B. Derudder, C.G. García and F. Witlox)

11. GaWC Research Bulletin 203 (Z) : 'The UK's Provincial Cities: Using Global Network Connectivities to Assess Claims of Economic Revival'
(P.J. Taylor and R. Aranya)

12. GaWC Research Bulletin 200 (B) : '“Gateway cities” - Los círculos bancarios generando concentracion y dispersion en lo sistema urbano de Brasil'
(E.C. Rossi and P.J. Taylor)

13. GaWC Research Bulletin 198 (A) : 'Where We Stand: A Decade of Empirical World Cities Research'
(B. Derudder)

14. GaWC Research Bulletin 194 (Z) : 'The Internationalization of Europe's Contemporary Transnational Executive Search Industry'
(J.V. Beaverstock, S.J.E. Hall and J.R. Faulconbridge)

15. GaWC Research Bulletin 193 (Z) : 'The Metropolization of the European Urban and Regional System' (S. Krätke)

16. GaWC Research Bulletin 192 (Z) : 'A Global ‘Urban Roller Coaster'? Connectivity Changes in the World City Network, 2000-04'
(P.J. Taylor and R. Aranya)

17. GaWC Research Bulletin 187 (Z) : 'Flying Where You Don't Want to Go: An Empirical Analysis of Hubs in the Global Airline Network'
(B. Derudder, L. Devriendt and F. Witlox)

18. GaWC Research Bulletin 184 (A) : 'Cluster Dynamics, Cluster Development and Innovation: Insights from Broadcasting in Three UK City-Regions'
(G. Cook and N. Pandit)

19. GaWC Research Bulletin 182 (Z) : 'Global City Frontiers: Singapore's Hinterland and the Contested Socio-Political Geographies of Bintan, Indonesia'
(T. Bunnell, H.B. Muzaini and J.D. Sidaway)

20. GaWC Research Bulletin 177 (A) : 'Cities within Spaces of Flows: Theses for a Materialist Understanding of the External Relations of Cities'
(P.J. Taylor)

21. GaWC Research Bulletin 175 (Z) : 'Transaction Links through Cities: 'Decision Cities' and 'Service Cities' in Outsourcing by Leading Brazilian Firms'
(E.C. Rossi, J.V. Beaverstock and P.J. Taylor)



* Marek Bańczyk, Tomasz Achrem, Jakub Mroz, Dariusz Śmiejkowski, E-mail:


Edited and posted on the web on 4th October 2007